On July 9, 2018, reader CoRev disparages futures prices as accurate predictors of future spot prices for soybeans, writing:
no one has denied the impact of tariffs on FUTURES prices. Those of us arguing against the constant anti-tariff, anti-Trump dialogs have noted this will probably be a price blip lasting until US/Chinese negotiations end. We are on record saying the prices will be back approaching last year’s harvest season prices.
My assertion was, based on econometric analyses in Chinn and Coibion (2014), the best forecast for the price of soybeans one year from date of forecast was the futures contract price for expiration one year away. So as of July 15, 2018, I would use the then closing price for the July 2019 contract: 872. (There is a subtlety that the contract expires on 7/12/2019, so really I should use the 7/12/2018 price of 884.75, but for some reason CoRev wants 7/15.) 872 is shown as a horizontal red line in Figure 1 below.
Figure 1: Spot soybean prices (black) and July 2019 soybean future contracts prices at close (blue). Red line at 872, July 2019 futures contract price as of 7/15/2019. Light green shading denotes price range 7/24-11/30/2018 (harvest period). Source: Macrotrends.com, ino.com, author’s calculations.
In contrast, CoRev hearkens back to the previous harvest season as of July 2018; that is late September to end November. I’ve shaded that period tan, and shaded the implied price range light green, in Figure 1.
Who came closer? The spot price on 7/15/2019 is 902, the futures implied price is 872, for a difference of 30 cents. CoRev’s range is 957 to 1000. The bottom of the range is 55 cents higher than the outcome. The average value during this period is 980.36, which is 78 cents above the actual realization on 7/15.
Recently, CoRev has tried to resurrect his forecast by after-the-fact revision, in this comment earlier today:
Here we are on close of the US Futures Market and the results are in. Today’s. close was $9.02 From here: https://markets.businessinsider.com/commodities/soybeans-price That shows that Menzie’s PREDICTION was WRONG!
That’s $.30 higher than Menzie’s prediction of $8.72 for this date from last year July 15, 2018 when the price was $8.30. The average trading price for the year was $8.56 (low $7.81 + High $9.31)/2= $8.56 The high price for the year was $9.31
My quantitative estimate was for the prices to APPROACH the 2017 market price as we approached the 2018 harvest season with a conditional only IF THE US/CHINA TRADE NEGOTIATIONS ENDED. They still have not ended. So the prices have been depressed at a new plateau since they tariffs were implemented. I picked early harvest to be the first months September and October, as any prices earlier would be based upon estimates and not actual harvest numbers. My guess without doing the analysis to add precision is the plateau is around $8.70-$8.80 for YTD since July 15 2018. Accordingly my conditional quantitative forecast should be considered with the new tariff driven plateau $8.70-$8.80 against the 2017 early harvest season (late Sept – Late Oct 2017) price average of $10.09 ($9.43 – $10.75) versus a similar period in 2018 $8.53 ($8.14-$8.92)/2.
My conditional quantitative forecast was wrong. Prices reached above that plateau price starting in November 2018, later in the harvest season not as we approached harvest, Sept and Oct.
I was off by ~$.17 -$.27. Menzie was off by $.30.
If this explanation for why CoRev’s forecast was superior to mine makes sense to you, please explain it to me.
CoRev…….. he’s kind of a stink bug if you ask me.
https://www.youtube.com/watch?v=oShTJ90fC34
See, stink bugs eat soybeans. [silence] I SAY [ !!!!! ] stink bugs eat soybeans. [silence] Aaaaaahhh!!!! Teh hell with yehs!!!!!
In that same 9 July 2018 thread CoRev asked me:
go on the record with a Winter price. Pick your own month.
And my reply:
Nov: $8.95/bushel
Your turn.
So let’s go to the board. The ending November price was…drum roll please…$8.95/bushel
https://www.macrotrends.net/2531/soybean-prices-historical-chart-data
Note that CoRev was perfectly willing to challenge others but demurred at making his own prediction. That’s how he’s always been.
2slugs, what part of the quote: “no one has denied the impact of tariffs on FUTURES prices. Those of us arguing against the constant anti-tariff, anti-Trump dialogs have noted this will probably be a price blip lasting until US/Chinese negotiations end. We are on record saying the prices will be back approaching last year’s harvest season prices.” is a demurral of my own prediction.
Please read and understand instead of emoting.
“this will probably be a price blip lasting until US/Chinese negotiations end. ”
Which you predicted would come quickly. It didn’t. Yea – I see where you never said the negotiations would end the trade war quickly (just today). But CoRev – you lied with that denial. Now I get it – Trump lies 24/7 so your lying 24/7 is seen as perfectly acceptable in Trump world. But that does not change the fact that you are a serial liar.
As usual, pgl, has not done any research to substantiate another unsupported claim.
He does amaze!
Research? Into what? How many lies you make a day? Seriously CoRev – I trust you know your own dog is laughing at you. He did not do any research either? He does not need to as your insanity is as plain as the food he eats.
CoRev Your quote of yourself was non-responsive to the question I asked. You asked me for a price prediction. I gave you one. It turned out to be right. You could have done the same. Instead you gave us hedging, equivocating and the usual CoRev bob-and-weave stuff. If you didn’t want to give us an unconditional forecast, then you could have given us two conditional predictions; one assuming the tariffs were still on and another assuming the tariffs were off. In July 2018 (when we had this discussion) the “last year’s harvest season prices” (i.e., 2017Q4) were in the $9.50 to $10.00 range. So you either have to admit that you were wildly off the mark in our prediction, or that you have to admit that you didn’t want to commit to a prediction that assumed the trade war would still be ongoing (i.e., demurred from answering the question).
BTW, your “price blip” is a peculiar term for a price drop. The term “price blip” usually connotes a temporary price spike, not an extended drop in price.
One last thing. You might want to take Menzie’s advice and learn something about statistical methods.
“Your quote of yourself was non-responsive to the question I asked.”
He generally does not respond to either factual statements or even good questions. It ain’t his style.
2slugs, I’ve learned most of what I know chasing your complaints. Frankly I thank you.
As to your conjecture, I could have done many things differently, but was unsure of the duration of the trade talks so made a CONDITIONAL forecast. Mostly due to your insistence they would continue longer than I thought. I have admitted several times that trade talk assumption was wrong. I’ve also admitted that the timing was wrong on my forecast. Prices approached 2017 prices later in Dec 2018 in the $9.12 to $9.20 range. That is approaching the 2017 harvest average $9.75 (using your price data). Remember we were dealing with the June lows ~$8.14 in those early July comments.
You seem to forget to apply my conditional: “… until US/Chinese negotiations end.” I hedged my forecast with my conditional and my imprecise estimate. Unless you can factor that in, making any statement of how i did is more than just problematic. Remember that’s how forecasts work.
CoRev: I think you are squirming too hard to try to evade your forecast failure. You said “blip” which in ordinary English language as I understand it means a short period of time — surely not a year.
CoRev is one of those low-IQ persons who are really convinced that every one else just is not smart enough to understand the point the person is trying to make. What CoRev fails to grasp is that everyone understands the point you are trying to make. It is just a really really silly (and wrong) point indeed.
CoRev’s own dog is smart enough to realize his master has been writing gibberish for a long, long time. I pity any dog who wakes up and realizes he is smarter than his own master.
Menzie, I see you are again changing the parameters of your 7/15/18 PREDICTION. I remember we agreed to use the https://markets.businessinsider.com/commodities/soybeans-price as a source for data. You also agreed to use the 7/15/2019 spot price from that source. Now you are going back to the INO source, which has no 7/15/20128 spot price. You are also trying to compare apples to oranges, my conditional forecast to your unconditional PREDICTION.
Only in the past few weeks have you attempted this your versus mine comparison. You now seem to want to play horse shoes, instead of analyzing your PREDICTION.
Top be clear my conditional, was that US China trade negotiations ended. They still have not ended, so I attempted to factor the conditional. If you don’t like that approach, your unconditional PREDICTION can not be compared to my conditional forecast. The only validation of your $8.72 PREDICTION is to compare against the $9.02 7/15/2019 spot price as you very simply stated in your original version:
“Conclusion
Soybean futures are remarkably good predictors of future spot prices of soybeans. Hence, my best guess of soybean prices one year from today is 872. …”
So when you ask: “but for some reason CoRev wants 7/15.) …”, it’s because that was your CHOSEN DATE, one year from today, 7/15/2018.
in the spirit of horse shoes, Menzie missed the stake and I have not yet thrown. Menzie’s PREDICTION was WRONG! Mine is indeterminate.
CoRev: I used whatever source I could get the exact time series data. I have spot rates for 7/15 from MacroTrends.com for 7/15/2019, I have spot prices for 9/24-11/30 to calculate average prices over the “previous harvest” you referred to in 2018. I get the futures price as of 7/15/2018 for the July 2019 futures contract (CBOT ZS.N19) settled in 7/16 (expired on 7/12). I don’t know if you understand any of these parameters surrounding the futures contract — your previous and recent exchanges suggest you do not — but I am using spot and appropriate futures contract prices.
You can get the MacroTrends daily data for free here: https://www.macrotrends.net/2531/soybean-prices-historical-chart-data and you can get the July 2019 futures contract daily prices here: https://quotes.ino.com/charting/?s=CBOT_ZS.N19
For ino.com you have to become a member to download daily data, but you can start a 30 day free trial.
So you have no excuse not to check my data and calculations yourself — except for possible incompetence or laziness (you will have to align both data sets to put them into 7 day/week format). Or are you going to accuse me of “lying” once again, as per your confusion about FRED data vs. BLS data?
Yes – finding spot prices is easy. And you have provided sources for future prices. But CoRev has no idea what either series represents even if we have tried over and over to explain this to him. I guess his inner ghost has drowned this all out with the nonstop “Trump is not a racist. Trump is not a racist”. Trump himself says he does not have a “racist bone in his body”. I believe that as bones themselves are just bones. The rest of Trump? Racist to the core!
Pgl huh!? Do you have any point? After months of making this similar claim: ” But CoRev has no idea what either series represents…” Menzie confirmed I was correct in my source for a spot price on July 15, 2019.
You do amaze.
Let me be plain since you asked. You are a liar. You are stupid. And you are a total waste of time. But thanks for playing!
😉
You do amaze!
I second pgl’s plain explanation. It is both brilliant and to the point.
Menzie, I did not question your sources of data. I too have access, but why all the obfuscation over the source? They are also available on https://markets.businessinsider.com/commodities/soybeans-price/usd including the data for YOUR CHOSEN DATE, and which you agreed to use as as the source.
Do you understand the parameters of your own PREDICTION?
“Conclusion
Soybean futures are remarkably good predictors of future spot prices of soybeans. Hence, my best guess of soybean prices one year from today is 872. …
This entry was posted on July 15, 2018 by Menzie Chinn. ”
Your best guess 872 and the actual $9.02 on July 15, 2019. You WERE WRONG. Obfuscation won’t change that.
CoRev: Regarding accusations of my not properly attributing the data, do you remember you writing the following?
That wrongful accusation is something you have never apologized for, nor even admitted you misunderstood that FRED data was in that case BLS data….
Stop digging.
Menzie, stop denying your PREDICTION was wrong and deflecting from it.
Continuing to bring up unrelated obfuscations are obvious.
CoRev: Well, all predictions are “wrong” in the following sense, if one is working with continuous variables (and soybean futures come close): there is exactly zero probability of hitting the exact price out to a million decimal places. Now, for soybean futures, we only go out to two decimal places, so it’s possible to hit exactly, but unlikely. So then, it becomes a question of getting “close” (i.e., we are in a horseshoes and hand-grenades world). In fact, that’s how the entire statistical-forecasting world works. And by that criterion, I am definitely doing better than you are.
Menzie, back handingly admits his PREDICTION was wrong: “… if one is working with continuous variables (and soybean futures come close): there is exactly zero probability of hitting the exact price…” Which I recognized from July 15, 2018. Why didn’t you recognize at my 1st comment, instead of changing tactic and starting this obfuscation campaign early in 2019?
Yet, he wrote a horribly phrased PREDICTION attempting/implying some precision and exactness in both date and amount. After quoting his exact phraseology many times he finally implies that’s not what was meant. Then write what you actually mean instead of some ego driven mistake.
After mocking all these months of mocking me for not understanding (a list goes here), you and your cohort still can not say the words. Use your words Menzie! 😉 An apology is appropriate.
CoRev: You *are* joking, aren’t you…In other words, unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.
Menzie, even pgl got that point several weeks ago. ” unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.”
I can’t help that you wrote a horrendous PREDICTION. What amazes me is you now claim to never have realized that? adding or at least referencing the uncertainty or conditions associated with it would have helped some, but your ego didn’t allow for that. For over a year you didn’t realize you were to be tested on your conclusion as written?
As I said in an earlier thread, I see these kinds of predictions in the climate world all the time. There it is deliberate to not show the u8ncertainty associated with them Even the IPCC, after being challenged, started showing the uncertainty/error bars and quickly changed their term to PROJECTION from prediction or forecast.
It is for this reason I continue to capitalize unconditional PREDICTION, which is what it is.
Like I said. You lie 24/7. And these are pointless lies. But do continue!
“unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.”
Yes. Wow. That is why CoRev, all of the rest of us think you need to go away. Your intellectual garbage is a waste of time.
Wow – the market price was a mere 30 cents higher? Lord – only the dumbest person on the planet would be shocked at such a small forecast error. Hey CoRev – no need to try so hard. EVERYONE knows you are incredibly stupid.
Hey calm down CoRev. No one believes that Trump is at fault for the Great Depression. And no one believes Trump’s bone are even racist. The rest of Trump – racist to the core. And incompetent at it gets. Sort of like your feeble defenses of his trade policy.
https://www.facebook.com/clickitconservativenews/photos/a.1725559317703597/2363557087237147/?type=3&eid=ARAoITCjfRWckWILA8hdcvQCymor9itS5Xs1ceAuQyzA-LjxW-ZjbqpesXQue0u5GsKwycx9q1hbcfB7&__xts__%5B0%5D=68.ARCOqVYKhAxwQQqH7evR_Okbs_ihB7ebIH53CIVRLhE2AMUDIK-JESIAuZxUpfXNpsNS5sibopv8DzrIF2oM78GCE6EDpAB4dRESRiMFY09_LlbknR_PvzWSW5qIesfE1DnTzXaNpx6zzOWdJX8Tq3RwcxlIyzZ5aKedRTF637FYKwx1yIzJngfGmbnwanA2UELmJo3EgP9u1tE-2xkF6fIzfKIMJLP4vpqRm7yKaIGs1bJSePYQXqDZZXOkwzuD3zJz9l9qV1-68FbZx_ohb41Cl2Yx63lZJCN3jy-UaKz5pJHrC4lhJOmLQqkIxPJF351CpqBt7hfRhEa8bubSDv0l3B6GFMW_BC4MbKeo4rvdALWohg&__tn__=EHH-R
LOL! You, Joe Kennedy, and all the Republican cowards in House. Trump clearly is a racist and anyone who cannot admit that has condoned his racism.
Menzie, you’re like the landlord in this old-timey TV show, you just don’t understand the new MAGA math:
https://www.youtube.com/watch?v=MS2aEfbEi7s
If it was up to you Menzie, white trash parents’ children would be forced to learn Common Core curriculum, and then people like CoRev would never graduate high school. Don’t you feel ashamed of yourself now Menzie?? It’s much better if we pass white trash losers so they can blame immigrants for the fact they can’t get a job. Plus they make for very literate voters that “don’t want the government to put their fingers all over my Medicare”. It’s a real sharp group we got out here “in the heartland of America” Menzie, and if you keep asking questions like the one you asked at the very end of your post we’re going to have to send you off to the MAGA re-education camp.
At MAGA re-education camp you learn important life morals, like when Republican leaders show congressional staff their foreskin it’s called “family values”.
https://www.youtube.com/watch?v=KVIDCSswc6E
https://www.youtube.com/watch?v=cvP3rm7JQjc
Professor Chinn,
Would you explain the spot price versus the futures implied price? I must have missed this concept in the several previous posts on this subject.
“The spot price on 7/15/2019 is 902, the futures implied price is 872…”
Who could blame you for not grasping this after all that incessant babbling from CoRev? Menzie is trying but the nonstop screaming from CoRev drowns out the conservation.
Is the 872 the implied price being currently cited the price from one year ago? If so, I see that from his past and current writings. I thought perhaps there may also be an implied current price from the futures that was being compared to the current spot price.
AS: There are four concepts: spot rate at time t, s[t]; futures as of time t for expiration on 7/15/2019, f[t,7/15/2019]; expected spot rate on 7/12/2019 based on time t information, Es[7/15/2019 | Omega[t], and spot rate on 7/15/2019, s[7/12/2019]. For my forecast for prices on 7/15/2019, I use the futures for the July 2019 contract seen on 7/15/2018; that means I am proxying the Es[7/15/2019 | Omega[7/15/2018] ] with f[7/15/2018, 7/15/2019] . The actual spot rate on 7/15/2019 was 901.75 , so my forecast error was 29.75. If I had used a random walk model (I think that’s what you were wondering about), I would’ve used the spot rate on 7/15/2019 s[7/15/2018] as my predictor for 7/15/2019. The spot rate on July 14, 2018 was 989 (there was no trading on 7/15), so *in this particular case* a random walk would’ve been superior to the futures. However, the entire point of Chinn-Coibion (2014) was to show that *on a repeated basis*, futures outperformed a random walk on a MSE basis.
There is a separate technical issue which appears to confuse CoRev. The actual expiration date for the July 2019 futures contract is 7/12, while the first settlement date is 7/16. For me, since futures prices can’t change after expiration, the proper reference date is 7/12. But since the contract price remains the same through 7/16, I guess 7/15 is still predicted by the 7/12/2018 futures price…
Menzie, I’m not confused at all. I’m evaluating your model on EXACTLY your phrasing for your best guess on 15 July 2018. That may not be what you intended to say, but after a year to correct you failed to do so.
It’s like pgl continuously showing a 45 year megatrends graph and not realizing he is showing weekly averages when the need was for daily prices. Yes, I’ve downloaded the source data from al the references sources. Some several times to respond to prior mockeries.
“I’m not confused at all.”
BAHAHAHAHA! You are the most confused clown ever. So confused – you actually think you are brilliant in your sheer and incessant stupidity!!!!!!!!!!!!!!!
Professor Chinn,
Thank you.
One additional clarification would help. On your blog entry titled,
” Guess the Expiration Price of July 2019 Soybean Futures” You said, “My guess is the expiration price will by 885.75”.
The 885.75 is only 16.00.
I may have fallen into the category of better to be silent and be thought to be a fool, than open mouth and remove all doubt.
Regarding updating the forecasting model, I had thought that using more recent data may be able to forecast closer ending price compared to the futures from a year ago, but alas, your year-ago forecast was similar to using updated data. Using recent data, on July 7, 2019, I forecasted a closing price of 870.82 as reported on a July 7, 2019 blog entry.
AS: I wanted to use the expiration date of July 12; the correct one-year ahead forecast then was July 12, 2018…
Got it finally. Thanks. I will move to the back of the class and be quiet.
AS: No, those were good questions, and in the welter of blog-commenting-back-and-forth, not clear (which is why blogs are not the ideal teaching medium — textbooks are…)
Ah, but here we have it. The problem is all those meanie engaging in “anti-tariff, anti-Trump dialogs.” If only they would stop that and get on board with how great those Trump tariffs are, those US-China trade negotiations would have ended long ago and the soybean price decline would have been just a “blip.” Shame on them.
And it is even worse than that. I mean, here it is on the half century anniversary of a great triumph for CoRev, the launch of Apollo 11 for which he received multiple nonexistent “Apollo 11 awards.” But when he claims to know more about climatology than anybody here Menzie has the temerity to ask him what is advanced degrees are, with us since learning that he has a degree in womens’ studies, with the diploma in a tube in a drawer. Not only that, the tube is in there with a medallion he earned that went on that glorious mission to the moon a half century, and returned safely.
Yes, it is truly shameful how terrible all these people engaging in anti-tariff, anti-Trump “dialogs” are.
😉
Barkley, as we approached the magical Apollo date I related my Apollo story on a couple of blog sites, and was surprised no other experienced personnel joined in. As you did on AB. I guess we Apollo vets may be like Korean War vets. We are getting past our expiration dates.
Wasn’t the expiration date for your brain 1969?
@ CoRev
Were you a janitor for Mission Control in Houston?? Can you tell us how difficult it was calculating the bleach to water ratio when cleaning the bathrooms for Gene Kranz?? I’m guessing that may have been a challenge for you. Maybe someday you’ll get a footnote in the history books.
When we start to fantasize about our own personal life history in very preposterous ways, it’s time to seek psychiatric treatment.
I’m up to page 72 in my “R” code learning book. The part that explains linear regressions doesn’t even begin until page 270. Have any betting pools started up yet on if I make it through this cr*p?? I figure between reading 2–3 books on this, memorizing the code to where I don’t have to keep using books or notes as a reference, and working through problem sets where I get at least semi good at it, I’m giving myself an 16 month window on this, roughly 1 month into the next term of presidency I have the goal of being able to crunch some of the numbers Menzie gives us in some of his easier examples. Although I still a little bit wonder if Python or Gretl would have been better choices, I’m thinking long-term “R” is the most suitable choice for me. I’m starting to feel semi-confident “R” is the correct choice for my personality and what I get gauged to. Most of the code command names have logic to me when they are explained, and if I can tie in meaning or tie in reasons for the code command names, then it makes it MUCH easier for me to memorize and commit to my mind. I really hate the feeling when command codes or anything I have to remember has a random or haphazard type feel to it. So this is one thing I like about “R” so far. And the problem I think I had with Gretl is there seems to be a limited community there, whereas with “R” there seems to be a relatively large community, so this is probably the biggest factor in me brushing Gretl to the side, at least for now anyway.
There is a sense in which CoRev is correct about the conditionality of being after negotiations ended. Then again, you can tell it’s in bad faith because he embeded the forecast that the negotiations would conclude in much less than a year, such that the futures price fall would be a blip, and is now trying to ignore that part of what he said.
IS, I said a blip IF US China trade negotiations ended. What is it that so few can understand what is written.
I’ve also admitted my expectation of them ending quickly and that my timing of the blip was wrong. When wrong I admit it.
“When wrong I admit it.”
You would have to say you’re wrong about 20 times a minute to live up to this claim. Let’s make this easier – let’s count the number of times you are right. Huh – zero it is!
“is now trying to ignore that part of what he said.”
Actually all of us should ignore all of what CoRev says!
I think CoRev’s message is “blame it all on Obama. Trump is the source of all goodness. And of course he does not have a racist bone in his body”.
😉
Trump much like his father, is a con man. His grandfather had a plan when coming here and his family has played race cards since then for 100 years. I also think Al Sharpton was a huge influence in learning how to race bait for the Trump organization. Lind and Strauss pretty much invented the modern Republican party playing the same kind of race baiting schemes. Controlling house slaves is fun. But in the end Vegas and Lukud run the show. The latter’s ties to the American Nazi Party and it’s financing via Permindex is one of the great conspiracies of the day due to the monied elite ignoring it on the internet cells. Controlling information is the key.
I wonder what the correlation between the spot and futures prices is.
From just eyeballing the data it looks like it could be 99.
So what do futures traders know that spot traders do not know?
I can see why the spread changes over time as futures and spot traders
trade for different reasons. What is the significance of the spread being
almost zero in recent months? Is that the market sending some sort of signal?
it is important to remember the context under which menzie is chastising corev. the argument basically began because menzie produced analysis which indicated farmers are directly taking a beating due to the tariffs and resulting trade embargo between china and the us, in particular on soybeans. menzie has shown that it is statistically relevant to consider trumps policies as directly leading to the depressed soybean prices of the farmers, and their struggles are a direct result of trump policy. now our boy corev, the ever defending trump acolyte, came to dear leaders rescue by claiming trump policy was NOT hurting the farmers. instead it was other factors (weather, oil, stray meteor impacts, etc) that were depressing the soybean prices. the trump effect was transistory, not meaningful, and would be overcome during the next harvest when prices would return to normal. after all, trump is a master dealmaker and he had a plan with respect to how all this would play out, and it would occur in the short term. dear leader is a genius businessman (having never run a business into bankruptcy) and we should follow his secret plan to fruition. over the past year, weather decimated the upcoming soybean harvest, which should have driven soybean prices higher. and yet………..they have not recovered. but trump trade policy is still in effect……and soybean prices are still lower. corev can quibble all he wants about spot versus futures prices, sources, etc. but don’t let him distract you from his position that dear leader had a plan in place, and dear leader would lead us back to profitability with his resounding wins in the trade war negotiations. according to corev, dear leader is not responsible for the financial beatings of the soybean farmers. its all about the weather.
That sums it up pretty well. Of course what makes me chuckle is how CoRev’s “bad weather” comes down to an inward shift of the supply curve. Yes in CoRev’s advanced econometric model, inward shifts of the supply curve are supposed to lower market prices. Go figure!
Baffled, don’t you read the articles? had you done so you might have seen the quote in the beginning: “no one has denied the impact of tariffs on FUTURES prices. Those of us arguing against the constant anti-tariff, anti-Trump dialogs have noted this will probably be a price blip lasting until US/Chinese negotiations end. We are on record saying the prices will be back approaching last year’s harvest season prices.” That was written even before Menzie’s PREDICTION.
BTW, also in context we have seen ASF in China and the US Spring planting season weather confirm my concerns. You also forget my viewpoint was through a farmer’s eyes, and not an academic economist’s. They are different.
Your making the same type of context mistake you made in defining “raw data” as electronic source. I guess you never read a liquid filled thermometer with temperature graduations. We were talking about temperature data if you remember. You’ve shown you believe this: “Sen. Hirono: Democrats Have a Hard Time “Connecting” With People Because Of “How Smart We Are”” and then they open their mouths or write a blog response. From here: https://www.realclearpolitics.com/video/2018/12/06/sen_hirono_democrats_have_a_hard_time_connecting_with_people_because_of_how_smart_we_are.html
Lord – you expect everyone to read the endless serial stupidity you pollute this comment section with? And you cannot even bother to read Fama’s seminal paper. The class clown turns unbelievably arrogant. No wonder you worship Trump. You are his minnie me!
“China’s trade surplus with the US rose 11 percent to USD 29.92 billion in June from USD 26.9 billion in May. For January-June combined, the country’s trade surplus with the US widened 5 percent to USD 140.48 billion from USD 133.76 billion in the same period in 2018.”
https://tradingeconomics.com/china/balance-of-trade
@ CoRev Can you remind us all how many months ago the racist creature named donald trump took office?? I forgot now and was hoping you could tell us since China’s bilateral trade surplus via the USA has been rising. I have another link here, can you tell which months since donald trump has been in office those numbers turned into a bilateral trade surplus for America via China?? I am having a devil of a time finding a figure that doesn’t have a negative sign attached to it.
https://www.census.gov/foreign-trade/balance/c5700.html
“You also forget my viewpoint was through a farmer’s eyes, and not an academic economist’s. They are different.”
core, the argument was soybean prices would return to normal once trump demonstrates his might and superior negotiating skills. your argument was this is a “blip” that will not last too long. it has lasted much longer than your “blip” argument implied. it is very clear to everybody (except some idiot on this blog) that the farmers pain is attributed to trump policies. not weather. or fungus. or stray meteor impacts. you simply need to acknowledge the damage trump has done to the agriculture community (among others). it is best to acknowledge that your fake farmers eye view was extremely flawed. perhaps it is wise to listen tot the experts, rather than assume you know more than they do.
your thermometer reading comment was simply stooooopid. quit ranting.
Baffled, I guess you don’t want to admit the ignorance of your raw data=electrical signal comment. “your thermometer reading comment was simply stooooopid. quit ranting.” i never made the claim it was YOU.
nobody is sitting in a satellite measuring a thermometer. again, your comment has no point other than a rant. it is simply a silly statement from an ignorant and pompous bloviator. go back and finish your beer cliff claven.
Baffled, for you surely are, the start of the temperature data sets ranges from 1850 -1880 (depending on the source). So saying this: “nobody is sitting in a satellite measuring a thermometer. ” indeed is simply a silly statement from an ignorant and pompous bloviator.
corev, once again, nobody is sitting in a satellite measuring a thermometer. satellite measurements require calibration and validation. but even your thermometer requires this same idea. you think you directly read the temperature? no. you read a device that requires calibration and verification for accuracy. we developed nist to address these types of issues. those gradation lines on the thermometer, they need to be calibrated for accuracy. it is not magic that the number reads 76 degrees. once again, at a fundamental level, you appear to have absolutely NO understanding of what raw data actually is, and the process required to make it into data others “analyze”. i sometimes wonder are you actually this stoooopid, or do you intentionally play the role of village idiot?
Your surely are baffled, because you are th e only one talking about the satellite data sets. My reference was to the OFFICIAL data sets of which the satellite data is a subset.
It’s OK! Amazing us is your mission in life.
once again, corev spews incoherent. i think you did answer my question, and the answer is village idiot.
baffling. That’s a pretty good summation. I would add that all of this is just a big set up for CoRev to claim victory when trump eventually capitulates with some face-saving agreement with China that doesn’t really change anything, everything does in fact go back to “normal” or close enough, and soybean prices finally do recover. See, NAFTA version 1.0001. The larger point being of course that none of this pain had to happen in the first place if Trump had any realistic and informed strategy for dealing with China. The other larger point being that Trade Wars are not easy to win and our tarrifs are paid by us.
I thought that NAFTA version 0.9. Or ‘US M C A’ sung by the Village People!
Dave still believes its all about the tariffs, and forget the ASF infection and the planting season weather: “…everything does in fact go back to “normal” or close enough, and soybean prices finally do recover. “
CoRev idiotically writes:
“I can’t help that you wrote a horrendous PREDICTION. What amazes me is you now claim to never have realized that? adding or at least referencing the uncertainty or conditions associated with it would have helped some, but your ego didn’t allow for that. For over a year you didn’t realize you were to be tested on your conclusion as written?”
The reason it is so difficult for intelligent people to grasp your (CoRev’s) argument is because it is difficult to believe someone could possibly have the brain power to actually type in complete sentences and then at the same time be so idiotic as to believe that the “argument” in question has any substantive merit. We are all guilty of desperately trying to glean some meaning out of the idiocy that spews from CoRev’s keyboard.
CoRev disingenuously writes:
“Dave still believes its all about the tariffs, and forget the ASF infection and the planting season weather: “…everything does in fact go back to “normal” or close enough, and soybean prices finally do recover. “
Dave has never said it is ALL about the tariffs. Why keep lying?
“Why keep lying?”
Easy to answer. His political masters pay CoRev by the lie. And CoRev has no other way to afford his rent.
“Dave still believes its all about the tariffs”.
Dave has never said that. Stop lying before your dog decides to bite you for this garbage!
‘Menzie Chinn Post author
July 17, 2019 at 8:55 am
CoRev: You *are* joking, aren’t you…In other words, unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.’
He is not joking. The other day he told us his forecast which was that the spot price would not be precisely what you wrote. In other words, if it was even slightly greater than your forecast, his forecast would be right. AND if it were below your forecast, his forecast would be right. Now THAT’s a BOLD forecast! Of course CoRev could not find anyone willing to take the other side of his idiotic bet, so he made precisely ZERO money on his bold prediction!
Dear Folks,
I am looking at this and am very confused. It would seem just as an onlooker, that the real way to settle this is to do it again. You both outline your conditions and how things change, and state what the spot price would be for 2020. I am not a CoRev fan, but am trying to be fair. It is not like there have been no tariffs announced, it’s just that the implementation is uncertain. So you say, “in my model, if we have a 10% tariff imposed on Mexican goods, then this is how the price will change, etc.” That way, the onlookers could see what in the model was wrong, instead of dealing with personalities.
Julian
You are being too tolerant, Julian. What happened was that in response to the tariffs Trump announced last year, China retaliated by blocking soybean (and other US ag) goods from entering China. This was followed immediately by a sharp decline in both spot ans futures soybeans prices. CoRev then leapt in with his prediction that this would be a blip because Trump would quickly get a trade deal, although he was a little vague about that. In effect Menzie said not likely, and the low futures prices are a forecast of that, with the trade war likely to still be in place keeping prices down. CoRev then went off about weather and much else, but in the end, which is what this thread is about, Menzie was right: the trade war did not end and today’s spot prices are not too far from what last year’s futures prices predicted they would be. CoRev has been trying to spin that bottom line away for various reasons, but none of them hold up well.
You just want to rerun fro next year what was already run? Really? Why?
Barkley, again, how many times do I have to admit that I was wrong about the tariffs ending sooner? Here it is again, just for you. I was wrong about the tariffs ending soon!
Now that that is cleared up, my complaint was about a really, really poorly phrased unconditional PREDICTION, which included a specific spot price, 872 on a specific date, 15 July 2019. Validation of this prediction was truly simple, and only the date was correct. Had he not meant that specificity he could have changed that prediction at any time during the year. Adding the uncertainty range would have been appropriate.
Instead, Menzie has changed the parameters of his original by saying I really didn’t meant that July 15 2019 date but, closure of the July Futures contract, July 12 2019, not that pesky July 15th. Which typically will change the price on a daily traded future contract.
Menzie also tried to detract attention from his estimate alone by making claims his was closer to mine. I did make a very conditional forecast based upon the mistaken belief that trade negotiations would soon end and a general quantification of price, approaching the 2017 harvest price. It is not possible to compare those two predictions without quantifying those soft areas of mine, the conditional, the period of comparison and term approaches. No one, but I, have done so to this date, when I admitted my timing was wrong. So claims of whose closest are not only deflections but wrong.
Instead of just re-writing his prediction into a forecast, Menzie for a year, he has doubled down on his really, really poorly written PREDICTION, which I recognized from the beginning.
Compounding this simple confusion was all the surrounding soybean articles starting even before the Menzie prediction article. Menzies’s early focus was on the trade issues. In June I made this statement: “Tariffs or other trade barriers/restrictions only are part of what makes up the price. As an economist you are ignoring annual supply and demand. ”
Well before that July prediction article I commented, using the farmers’ viewpoint that weather and other influences were a constant driver in their crap shoot of a livelihood. Then came the 2019 US Spring planting season WEATHER where a large number of farmers could not even plant their fields. That probably raised the soybean futures prices. Good weather was also instrumental in making the Brazil soybean harvest to be its 2nd highest allowing ample alternative sources for the Chinese buyers. Alternatively the African Swine Flu infection in the Chinese hog herd has suppressed demand causing lower prices, all supply and demand issues.
Many here misunderstood or allowed their own biases to misread what I had written. Through this old man’s eyes its a negative reflection of their objectivity.
CoRev,
Thank you for confirming that you were wrong about tariffs ending soon. I have no comment on the rest. Not worth bothering.
Julian. CoRev has no model. This is not a fair comparison. CoRev has finally explained that the sum of his “prediction” entirely and at all times was “I disagree.” And, CoRev now claims victory for his “prediction” .
CoRev pretends to have a model. Alas, his model is so incredibly screwed up that not even Lawrence Kudlow would endorse it.
His preferred model is Angie Dickinson.
CoRev Menzie, even pgl got that point several weeks ago. ” unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.”
In the blogosphere you have long had a reputation for intellectual dishonesty. The stupid point you’re making here is just further confirmation of that dishonesty. Even you aren’t that dumb to not know that all predictions are bounded by error terms. That’s something that most people learn in grade school. What’s worse is that if you had bothered to read the paper on soybean predictions you would have seen an error term in the prediction model. Should we count that as yet another example of intellectual dishonesty or would you rather chalk it up to intellectual laziness (a.ka., the heartbeat of Trump-era conservatism)?
What amazes me is you now claim to never have realized that? adding or at least referencing the uncertainty or conditions associated with it would have helped some
Well, I guess it’s sheer intellectual laziness because the uncertainty was right there for everyone to read. I’m sorry if you’re the only one here who couldn’t be bothered to read the paper. Are you asking Menzie to set up a special ed class for the resident slow one here?
I see these kinds of predictions in the climate world all the time. There it is deliberate to not show the u8ncertainty associated with them Even the IPCC, after being challenged, started showing the uncertainty/error bars and quickly changed their term to PROJECTION from prediction or forecast.
Are you trying to convince us that if they did explicitly show prediction intervals that you would even have a clue as to how to understand them??? Please, give us a break. You don’t even know the difference between a trend stationary and a difference stationary time series.
“CoRev
July 17, 2019 at 10:44 am
Menzie, even pgl got that point several weeks ago. ” unless I had precisely hit on the mark out to two decimal places the price on 7/15, I was going to be WRONG…? Wow.”
I can’t help that you wrote a horrendous PREDICTION.”
Mother of pearl. I was mocking CoRev. And this serial idiot say I got the point. Yea I did but the moron extraordinaire had no clue how badly he was just mocked that he actually repeats my mocking as if it was confirmation of his dishonesty.
DUMBEST. TROLL. EVER!
2 slugs, yes i do and would understand them, uncertainty and trend/difference stationary.
In the blogosphere blah, blah .. just another attempt to mock. Why? I do know more than most here about climate, and certainly more than you. You were the one who wanted to take the seasonality out of ANNUAL AVERAGE temp data. Bleh! in the blogosphere you have been called arrogant, and you do epitomize this: https://www.facebook.com/clickitconservativenews/photos/a.1725559317703597/2363557087237147/?type=3&eid=ARAoITCjfRWckWILA8hdcvQCymor9itS5Xs1ceAuQyzA-LjxW-ZjbqpesXQue0u5GsKwycx9q1hbcfB7&__xts__%5B0%5D=68.ARCOqVYKhAxwQQqH7evR_Okbs_ihB7ebIH53CIVRLhE2AMUDIK-JESIAuZxUpfXNpsNS5sibopv8DzrIF2oM78GCE6EDpAB4dRESRiMFY09_LlbknR_PvzWSW5qIesfE1DnTzXaNpx6zzOWdJX8Tq3RwcxlIyzZ5aKedRTF637FYKwx1yIzJngfGmbnwanA2UELmJo3EgP9u1tE-2xkF6fIzfKIMJLP4vpqRm7yKaIGs1bJSePYQXqDZZXOkwzuD3zJz9l9qV1-68FbZx_ohb41Cl2Yx63lZJCN3jy-UaKz5pJHrC4lhJOmLQqkIxPJF351CpqBt7hfRhEa8bubSDv0l3B6GFMW_BC4MbKeo4rvdALWohg&__tn__=EHH-R
If you want to live up to your arrogance, re-write Menzie’s PREDICTION. You have 3 things with which to work, 1) the model, 2) the date(s), 3) the spot price as shown in the original article.
“2 slugs, yes i do and would understand them, uncertainty and trend/difference stationary.”
Really? I guess there are multiple people posing as CoRev? You should ask those other CoRev’s to cease writing really stupid sh#t.
CoRev . You were the one who wanted to take the seasonality out of ANNUAL AVERAGE temp data
Your memory is failing. A few months ago you admitted that you misunderstood. I’ve told you several times that I used MONTHLY temp data and MONTHLY CO2 data. You need to start taking notes.
Hah, ha, ha, That’s so funny I have tears in my eyes. Your denials of you ignorance are getting worse. 😉
2slugs, your denials of your ignorance are getting worse. I’m laughing so hard I have tears in my eyes. 😉
BTW, it was my data you to which you were referring. Not yours.
CoRev in the blogosphere you have been called arrogant, and you do epitomize this:
One of the few things you got right.
Uh-ph….. hit my first snag on “R”. Wasn’t able to install “ggplot2” which should be a very easy process. I suspect this is because I was too lazy to update my “R” studio version and this is most likely the problem. We’ll see later today when I update my “R”studio version. If that is not the problem then I will be building up to a very serious conniption fit in the very short term time horizon.
Well, I don’t use “R” studio. I have it, but I’m a dinosaur and prefer the basic “R” (version 3.6.1). Then again, back when cavemen roamed the earth I also preferred the old dBase 3 to the later GUI versions. You should be able to select the Install Packages option from the “R” Packages menu at the top, then select a nearby mirror. Or you could go out on the web go to the site for ggplot2:
https://cran.r-project.org/web/packages/ggplot2/index.html
and save the zip file. You can then install the zip file from “R”. The ggplot2 site also has some useful vignettes.
Updated…….. successfully installed ggplot2 into Rstudio. [ Starts jumping around computer room uncontrollably yelping out “Oy!!!! Oy!!!!” like Rick James singing “Superfreak” ]
https://youtu.be/slOehjhqvk4?t=113
One I other thing. If you install from the zip file, make sure that you have all of the dependencies installed on your machine. The ggplot2 website lists all of the packages that are dependencies for ggplot2. If you install from the dropdown menu, then “R” should automatically load all of the dependent packages for you.
@ 2slugbaits
Appreciate the “heads up”.
Yeah, I noticed that, I did it from the dropdown, and both times, the successful install and the failed install it did that (had some text related to dependencies and where it was getting the “source”). The humorous part was when all of that stuff was scrolling/flying down in the Console portion on the left (and on the 2nd attempt it was actually much longer) part of the text said “lazy loading” and I was halfway wondering if the programmers put that in the code for people like me who were too lazy to update the version they were using. Actually I think it was unrelated, but it was kinda funny that that was why ggplot2 didn’t work on my first try. I was pretty excited to see that ggplot2 had installed, because I had gone down this avenue with “R” before, and one of the reasons I quit so easily was I couldn’t get any of my installs to work. Maybe this was like a year and a half ago?? But I just got so exasperated, and I think it was on installs I just gave up. It’s a “baby step” in actuality doing those installs, but to me it was a big deal. Because now I know I can get all those other programs. One is called “car” I think, and it’s tied into a book. There’s two big books that are tied into this stuff, and eventually I want to get around to those—you are probably already be familiar with those names. I will look them up again and put them very near to this comment if you are curious later. But to be a “jedi_Master” at this stuff, I really want to get into those two books—but that’s WAY down the road for me.
@ 2slugbaits and Frank
OK, these two books might be of interest to you, if you use “R” at all or know anyone who uses “R” and they might even be of interest to Frank if he’s reading from around Valdosta way there. Now I should stress this is for those more in the latter stages of their learning “R” than in their early stages of learning like me, because these are more advanced texts. But the two ones that look super good are “Kleiber and Zeileis 2008” and “Fox and Weisberg 2011” and I think Professor Chinn may have mentioned Wooldridge before (hopefully I’m not misquoting him on that). I think Wooldridge is more steeped in Econometrics as pure Econometrics, but does have an appendix related to “R”. So I’m hopeful that might give commenter Frank, or YOU 2slugbaits some kind of assistance on your learning journey.
Hello Moses,
Thank you very much for the reference. I, indeed, will try to add them to my library. They are the kind of textbooks I would be interested in.
Cheers,
Frank
you are a brave man moses. personally, i stick with matlab rather than try to learn R. although real men use maple. my better half used R a while back for some analysis and plotting, and it was very good, but limited to copying others code and running on the new dataset. learning R from scratch was a bit too much at the time. that said, if you really want to immerse yourself in frustration, try learning the new quantum computing codes put out by microsoft and ibm. i can now write a simple binary addition code to run on a quantum computer-and check the results faster on my 20 year old casio calculator! i guess there are worse things i could do to pass the time…
@ baffling
I greatly appreciate the compliment, but better withhold the “brave” until I can exhibit linear regressions, linear smoothing, and present a box that matches all of Menzie’s t–p—and z scores. I tend to “flame out” on these little projects. My hopes are higher though since I am taking a liking to the terminology of the code and my installs are working well recently. So, I’m pretty sure I chose the right route, and it’s going well so far. Just have to be persistent. My Dad always said people getting their advanced degrees was more a function of persistence and “doggedness” than intelligence. It’s obviously far below getting an advanced degree but I think that is kinda gonna be my personal slogan on getting where I can do “R” in a solid fashion.
I just got an error I have the strange feeling I am going to be seeing 5 billion times over the next few months. “unexpected string constant”
Be sure to pray for Uncle Moses kids.
Did my first scatterplot in “R” tonight. OK, it was from data already in the library, and, OK, it was code I was just copying from a book—but it did squirt out a scatterplot on my screen when I hit Enter, so, that is pretty cool.
All my homework in “R” paid off in a happy dessert at the end of the day. Slurp slurp!!!!
https://www.youtube.com/watch?v=aApyiK-dDtc
Anyone up for some Lobongo Latika!?!?!?! Huh?? What did you think I was imagining slurping on??? You bunch-uh damned sickos!!!
Dear Folks,
OK. Let me address a question to CoRev directly, then. Do you have model of futures prices, any of whose characteristics you are willing to discuss publicly? Or are we back to the situation with Montagu Norman – as in https://en.wikipedia.org/wiki/Montagu_Norman,_1st_Baron_Norman – where it is just a matter of intuition, and only results, regardless of chance variation, can be discussed?
Julian
Julian, I am neither a statistician nor an economist. My forecast was a matter of experience and and historical price analysis with some intuition, but that is true for almost every model. The biggest difference is the difference in our experience as shown by the trend analysis we both did for labor, and for which I was mocked.
As an example of the different experience, Menzie commented about the existence of some snapshot data, for which I showed him the snapshot on my inactive blog showing the Hiatus in temperature rise in the official temperature data sets, another data occurrence he has mocked. BTW, is the trend determination methodology often used in climate science.
I don’t know what you are asking with this: “… only results, regardless of chance variation, can be discussed?” In an unconditional PREDICTION, as Menzie provided, what else could be evaluated/discussed? Again, in my experience of the Climate field, these types of predictions are discouraged or expected to include error bars or some uncertainty measures. That is because of these kinds of discussion that have been held over the years.
The reason for those discussion is the same in both fields. Huge amounts of money are/can be expended on the assumptions of model prediction precision. When not shown false precision can be costly. If as 2slug and Menzie have insisted Menzie’s example of PREDICTION is common in economic, then it is a very poor practice. He should alter his teaching instead of continuing a poor practice.
“official temperature data sets” and this comment completely explains why corev fails to understand data analysis and interpretation. he has chosen an official data set magically. now this data set was produced by correlation and validation process, but he cannot conceive of the idea that perhaps it needs to be updated once again as we better understand the measurement processes. but it is the “official temperature data sets” for all of mankind and for all of times…
Baffled the issue was there a hiatus showing in the official data sets of that period and those versions? It was just recently reported that the UK version of its World temperature data titled RAW is a renamed version 3. Where did that raw data go? Why would any scientist rename the processed ” now this data set was produced by correlation and validation process” as the RAW data? It, the then processed version 3 data, was bad enough that it’s already replaced with even more processed data.
Sometimes the liberal mind does amaze.
again corev, this is nothing more than incoherent rambling. if you would like a valid and fair response, make a coherent argument. the fact you cannot do so is because you do not understand the process of collecting field data and turning it into something that other analysts may use. you are simply willfully ignorant of the process, and this frees your mind to continually make incoherent and stoooopid comments.
CoRev Just so we’re all understanding, it sounds to me like you think UK-Met, NOAA and other similar agencies should just do a braindead dump of unscrubbed data irrespective of errors. And presumably highly qualified experts such as yourself can then apply advanced data scrubbing techniques to correct errors. Have I got that right? Is that your understanding of “raw” data?
2slugs asks: ” it sounds to me like you think UK-Met, NOAA and other similar agencies should just do a braindead dump of unscrubbed data irrespective of errors.” Yes! They do and/or should make that level of data available for others to process in their unique ways. More importantly that level of data should be available to replicate those aggregation agencies processing. It’s a significant part of the scientific process
That is an excellent description of raw data. Raw data can also include the minimum of processing to either identify or replace/repair errors.
The temperature data usually shows an error or incomplete field by filling with all 9. However that is not the problem I described with the UK data.
BTW, NOAA is the final holder of all temperature data. If you’re interested this directory will get you processed data. https://www1.ncdc.noaa.gov/pub/data/ghcn/v4/documentation/
This guy did a comprehensive review of GHCN versions 3 and 4. This reference is just a preliminary look at setting up problem for GHCN V4: https://chiefio.wordpress.com/2019/03/18/ghcn-v4-first-peek/
If you want to be enlightened about the data review his several months of findings.
2slugs, as you can see corev really has no understanding of what “raw” data really is. if he actually received the raw data he desires, it would be filled with ones and zeros, of which he could not comprehend anything. but yet in his mind he could process it and produce more robust results than what we see by noaa and nasa today. corev thinks raw data appears in a formatted excel file downloadable on the web.
You surely are baffled, not excel but a TAR file format. Go ahead and download those ones and zeroes and be amazed by the nines filled fields. If you don’t know how to do that download and unpack use a Linux box and it will be automatic, and be even more amazed. You’re on your own after that.
Amazing us is your forte.
Dear CoRev and Others,
I am asking about models because a specific feature of the models can be proved or disproved by experience, and can be replicated by others.
A bad model of the soybean prices, but one that could be analyzed, would be
Y_t = soybean spot price = a (a constant, that can be statistically estimated, or just assumed to be a certain number) + b (another constant, same thing)*x_t (length of time a tariff is imposed on certain goods) + e_t (random error)
So we could look at the varying lengths of time a tariff was in place, and estimate a and b. Statistically, the characteristics of e_t would also be estimated, and one could make arguments for what the distribution of the random variable e_t would be. If there were a nuclear accident in China, that could be considered a random event, and it might be such as to be characterized by the assumed distribution for e_t or not.
Again, CoRev is arguing for a specific type of trend analysis. This would generate the coefficients for a and b, which could be examined. If I understand CoRev properly, his trend analysis for labor is deterministic, and is estimated in a very similar way to the model above. Most economists are skeptical of such trend analysis, even when they use it, as I admit to doing, and the whole idea of the Johansen-Juselius tests are a way to evaluate such trend analyses specifically. (I was very uncomfortable with Soren Johansen when I encountered him, and liked Katarina Juselius much more. But that may be similar to disliking Trump and liking Pence, or the other way around – both groups have each partner linked to another.)
This sort of procedure is what I have taught introductory students before. So it is a way to look at things without saying “I am right because I have experience in the field”. A similar example is whether the data in Compustat are truly representative of all U.S. businesses or not – you can do a statistical model of something you want to check and find out, by getting a survey or census. And before people laugh and say a census is impossible, the Canadians have done it for Canadian businesses, and it was reported in a very useful book by Baldwin and Gellatly.
In so far as I understand it without having reviewed the paper in depth, Menzie’s forecast was a conditional forecast, because he was making certain judgements about how long the tariffs would last. If the tariffs lasted to the exact second he predicted, it would still be a conditional forecast, because he was assuming certain other things would be constant. They may not have been. If we can get to whether the other things were really constant, and not argue exclusively about the past tariffs, especially since they clearly are not the last that will be imposed, then maybe we can make some progress.
Julian
I don’t understand the attack on Professor Chinn’s model. As he said above related to his article, , “the entire point of Chinn-Coibion (2014) was to show that *on a repeated basis*, futures outperformed a random walk on a MSE basis.” On any given month, the random walk model may outperform the futures model, and perhaps on any given month any other forecasting model may outperform the futures model.
I think that Julian Silk was trying to be fair in his questions and wants to know if you have a forecasting model. As you say, you used “…experience and historical price analysis…” It would be interesting to know what type of historical trend analysis you used.
As Professor Chinn says on page 610 of his article,
“The notion that the futures price is the best forecast of spot price is an implication of the efficient market hypothesis. If this is true, then price patterns are random, and no system based on past market behavior can do other than break-even.”
“The spot/futures pricing relationship is based on the assumption that market participants are able to trade in the spot and futures markets simultaneously, that is, they can utilize spot/futures arbitrage.”
AS, I know of no one who attacked Menzie’s model. What I challenged was the actual PREDICTION as written on July 15, 2018 that came from that model. It was the accuracy and precision IMPLIED with that phrasing. As you showed above, even Menzie was not trying for the level he implied: “the entire point of Chinn-Coibion (2014) was to show that *on a repeated basis*, futures outperformed a random walk on a MSE basis.”
AS, I didn’t do a trend analysis, but a historical analysis. I averaged the May/June average 2017 and 2018 prices. I was looking for a difference to estimate the 2018 early harvest prices assuming the trade talks had ended. Also in looking at the trend charts noticed consistent price drop at that same time of year for the past 2-3 years.
My logic was if the trade talks end 2018 prices will rise, and in picking the period I did I was comparing to a couple of past low periods. Remember my forecast was for the 2018 early harvest season.
CoRev,
Thank you for the response and the description of your approach. I think this may qualify as a model, which is what Julian Silk was asking about. I leave it to the experts to determine if your model is subject to MSE or other statistical measures of accuracy.
“assuming the trade talks had ended”
the issue most folks had on this blog, corev, was that you were arguing with certainty that trade talks would end soon, and trump would win the battle, resulting in a return to past prices. dear leader is a master negotiator. what most folks on this blog argued was that this assumption was poor. there was no indication the trade spat would end soon and produce a positive outcome for the farmers. you would not accept this possible outcome, to protect your dear leader. and yet this is the outcome we have gotten. so lets repeat one more time. trump trade policies have significantly hurt the farmers of america.
Baffled, and I have admitted I was wrong about the trade talks. How many times do you want to beat that dead horse? Maybe until those hand written glass tube thermometer derived temperature data are converted to one of your electronic instruments instead of until just converted to digital form for computer processing as they have been.
You do amaze.
AS, the underlying issue to my forecast is the need to carefully word it. You probably have noticed that the few conservatives still commenting here are relentlessly attacked, and any poorly phrased comment is pounced upon. Lately that has included Menzie as an attacker.
That is why I was surprised by Menzie’s phrasing of his PREDICTION. No mention of uncertainty or description of conditions made it an almost certain failure as soon as written. My experience with these kinds of predictions/forecasts/projections is in Climate Science where it was resolved early on due to discussions similar to the past soybean saga.
Climate scientists seldom predict as did Menzie, but now project future conditions with some attached conditions and uncertainty. When we do see these kinds of uncertainty missing claims it is almost surely in the news organs and not in the actual scientific papers cited. These types of claims are then just click bait. I have been chastised many, many times by calling a climate change projection a prediction. So I am sensitized to the differences when forecasting specific, unconditional and certain quantitative events. A climate science article ALLUDING to the difference: https://www.americanthinker.com/blog/2008/12/ipcc_we_dont_predict_we_projec.html
An IPCC definition of its use: http://www.ipcc-data.org/guidelines/pages/definitions.html
From that IPCC link:
“Forecast/Prediction
When a projection is branded “most likely” it becomes a forecast or prediction. A forecast is often obtained using deterministic models, possibly a set of these, outputs of which can enable some level of confidence to be attached to projections…”
Use of deterministic and stochastic modeling differ in our chosen fields of interests. Is that worth a year+ of mocking articles? Dunno. He espouses unquestioning belief of a major component of climate science using statistical techniques he disbelieves. Is he more correct in one belief versus the other? Again, I dunno.
The bottom line Is there one best method? Is using a possibly error prone complex modeling method better than a simpler method? Does writing an unconditional, without defining uncertainty a valid PREDICTION? Validation against the real data on the specified date says NO! To have been as accurate as implied by his form and phrasing would truly have been a random event.
That was a given from the minute it was written.
CoRev Since you seem to have expressed an interest in using strictly correct terms, let me add to your education.
When we do see these kinds of uncertainty missing claims
There is a well established difference between variability and uncertainty. The kind of thing you’re referring to (e.g., error bars, prediction intervals, etc.) are examples of variability. Variability is quantifiable. You can measure the first and second moments. Uncertainty is not quantifiable. With uncertainty you not only cannot quantify the second moment, you can’t even quantify the first moment.
These types of claims are then just click bait. I have been chastised many, many times by calling a climate change projection a prediction.
Perhaps, but not here. Some of us chastise you for lots of things, but I don’t recall anyone here ever chastising you for calling a climate change projection a climate change prediction. No one would waste their time on such a distinction because your “projection” or “prediction” would have been just as wrong regardless.
Use of deterministic and stochastic modeling differ in our chosen fields of interests. Is that worth a year+ of mocking articles? Dunno.
This is nonsense. The decision to use a deterministic or stochastic trend model isn’t a matter of differing tastes across disciplines. They are very different models with very different assumptions about the underlying data generating process and have very different forecasts. If all you are modeling is the trend in temperature anomalies, then the data are not mean reverting and you cannot reject a unit root. If you don’t believe me, then do the analysis yourself. For example, take the annual GISS data. If you know anything at all about time series analysis it shouldn’t take you more than three or four minutes to quickly conclude that an ARIMA(0,1,2) w/constant drift term is a pretty decent model in terms of the usual diagnostics; viz., normality of residuals, ACFs showing only white noise in the residuals, etc. A deterministic trend model fails almost every diagnostic test and performs lousy out of sample. Of course, a stochastic trend model does not pretend to model the physical reasons for climate change, only the observed behavior of temperature.
“Baffled, and I have admitted I was wrong about the trade talks. ”
corev, what you need to admit is that you were wrong about the ability of dear leader to “win” in his trade battle. you need to admit that trump trade policy has been brutal to the american farmer, especially the american soybean farmer. their struggles are not a result of weather, fungus or random meteor impacts. their struggles are a direct result of trump policy. you cannot seem to admit that item.
2slugs, I think you are confused again: “There is a well established difference between variability and uncertainty. The kind of thing you’re referring to (e.g., error bars, prediction intervals, etc.) are examples of variability…Uncertainty is not quantifiable.”
This is my understanding and use: “Uncertainty is quantified by a probability distribution which depends upon our state of information about the likelihood of what the single, true value of the uncertain quantity is.( You claimed it was unquantifiable.) Variability is quantified by a distribution of frequencies of multiple instances of the quantity, derived from observed data. That both are represented by ‘distributions’ is a major source of confusion, which can lead to uncritical adoption of frequency distributions to represent uncertainty, and thus to erroneous risk assessments and bad decisions. For example, the variability of natural phenomena is sometimes well-approximated by normal or log-normal distributions, but such distributions may not be appropriate to represent the uncertainty in outcome of a single occurrence.
We show there is no objectively ‘right’ probability distribution for quantifying the uncertainty of an unknown event – it can only be ‘right’ in that it is consistent with the assessor’s information.( on july 15, 2019 and an $8.72 spot price.) Thus, different people (or teams or companies) can legitimately hold different probabilities for the same event. Only in very restrictive, arguably unrealistic, situations can we choose to use a frequency distribution derived from variability data as a probability distribution to represent our uncertainty in an event’s outcome.
Our experience as educators of students and oil & gas industry personnel suggests that significant confusion exists in their understanding of the distinction between variability and uncertainty. ” From here: https://www.onepetro.org/conference-paper/SPE-169850-MS
Can you define either uncertainty or variability with Menzie’s CONCLUSION/PREDICTION as written?
No, I never said it occurred here. Econbrowser has a different form of attack, personal and mockery.
Your 3rd point was related to statistical models: “…then the data are not mean reverting and you cannot reject a unit root.” We are talking global temperature data. How often do we need to make these determinations on this old data? We’ve had this discussion in the past, and you have NEVER answered that question.
If you don’t like the Climate Scientists process then take it up with them. How often do good statistic practices need to be run against the same data? Maybe that’s why it takes larger and larger super computers to process the models. 😉
CoRev You copy & pasted something you found in a Google search without actually understanding a word of it. The authors’ distinction between uncertainty and variability is exactly the distinction I was making. When they refer to quantifying in the context of uncertainty they are talking about people assigning subjective range estimates. People do that kind of thing all the time. As the authors pointed out, quantifying for variables comes from the frequentist approach; e.g., you can observe and estimate the first and second moments using historical, frequentist data.
Can you define either uncertainty or variability with Menzie’s CONCLUSION/PREDICTION as written?
Yes. Read the paper. It’s literally right there in black and white.
We are talking global temperature data. How often do we need to make these determinations on this old data? We’ve had this discussion in the past, and you have NEVER answered that question.
It’s a stupid and unintelligible question. No matter which data set you use the temperature anomalies will have a unit root.
If you don’t like the Climate Scientists process then take it up with them. How often do good statistic practices need to be run against the same data?
Actually, I have. I’ve done some time series analysis guest lecturing at UAH & Redstone for some of Dr. Roy Spencer’s grad students.
How often do good statistic practices need to be run against the same data?
Good practices? Only once. Bad practices? As long as it takes.
Maybe that’s why it takes larger and larger super computers to process the models.
You don’t need super computers to run a simple time series model that describes the behavior of temperature anomalies. The super computers are needed to describe the processes buried in the drift term. If your goal is to come to a deep understanding of how various forces interact with climate, then you need super computers and lots of messy math. But if all you want to do is describe the time series path of temperature data, then a simple time series model is all you need. If you want to get a little fancier, you can find a cointegrating relationship between the nonstationary climate forcing variables (see James Stock’s paper on this).
2slug, nope! I think you are still confused.
I see you can not: “Can you define either uncertainty or variability with Menzie’s CONCLUSION/PREDICTION as written? ” You also demurred when asked to re-write his conclusion. It’s all there, but I guess you don’t want to show up the Prof.
I see you’re still referencing your visit down the UAH and Dr Spencer. He made an impression on you. Did he implement any of your suggestions, or were you doing comparisons processes of your two fields?
CoRev: Is your objective to embarrass yourself in front of the entire statistics community? Nobody I know of in the forecasting community (who’s had basic statistics) says “I predict GDP growth next quarter to be 2%”, and then expects to be evaluated as WRONG (i’ll put it in ALL CAPS as that is how you seem to like conducting a debate) when growth turns out to 2.1%. Instead, most people would say one is “more wrong” when growth turns out to be 2.3%, and yet more wrong when it turns out to be 2.7%; but perhaps as wrong at 2.3% as when it is 1.7%…
Now, I could tweak this metric, let’s say by allowing penalizing downside more than upside, or by using a discrete function (1.9-2.1% bad, 1.7%-1.89% and 2.11-2.3% equally bad), etc. But the general point is it’s not usually a penalty function that says “hit 2%” or otherwise “wrong”.
If you want to pursue your approach further, and make yourself the laughingstock of the internet, I am happy to oblige by writing an entire post on prediction evaluation. If you would like to put your real name to the approach, I am happy to further oblige. I can also post a link to your “anomaly analysis” for the sake of completeness.
I look forward to your reply at your earliest convenience.
Menzie, the difference in our experience is showing. It is a frequent complaint that climatologists are poor statisticians, but still they continue with the practices with which they are accustomed. The 1st email I ever sent you, many years ago, was to query the use of a specific test you had referenced. You responded with – I dunno mebbe.
Climatologists have altered their forecasts to add the uncertainty and conditionals associated with them, because of arguments similar to what I have made to yours. That false certitude can cost billions, or as in the Green New Deal, trillions.
This study: https://web.stanford.edu/~mburke/papers/climate_uncertainty.pdf is provided to explain the need for including or better handling UNCERTAINTY in climate studies and why. Let me clear. I think it, like so may similar climate studies, has a fatal flaw. It concentrates on on the negative impacts with little to no acknowledgement of the existence of the positives .
This paper says this about climate point estimates, similar to your on July15, 2019 will be $8.72.
“We show that accounting for climate uncertainty in these studies is both statistically and economically important. Doing so yields different point estimates, a much wider range of projected impacts, and much more negative “worst case scenarios”. (Notice the emphasis to which I complained above) In particular, failure to account for climate uncertainty can underestimate the range of potential climate impacts by a factor of five or more, relative to an approach that only considers uncertainty in the historical relationship between climate variables (such as temperature and precipitation) and the outcome of interest. ”
The introduction includes this:
“…Results of these studies have featured prominently in public policy debates, informing decisions about appropriate investments in greenhouse gas emissions reductions as well as in measures designed to help societies adapt to a changing climate. Such investments represent potentially large amounts of resources…”
I consider your “point estimate” to be a bad practice for economists as well as climatologists for the same reason they give false impressions of precision, and both can impact large tax payer resource allocations.
To explain my complaint about the “worst case” focus of this and many, many climates studies, there was almost no attention given to the positive cases also prevalent. For instance, previous warm periods during this interglacial have been given names associated with societal expansions, Minoan, Roman and Medieval Warm Periods. Incidentally they each were warmer than today’s, and generally lower than the previous. Yet, we continue to see this “worst case” focus without considering this well known and accepted climate history.
Menzie, let me take another approach, than the experiential, one above to explaining my position.
I view climate and economic data as MOSTLY random walk with drift, and we are concentrating studies on identifying those random factors and their impacts. Accordingly, since many of these random factors and their impacts are not yet known, then unconditional point estimates without uncertainty are both risky and will almost surely be wrong. As was your July 15, 2019 and $8.72 best guess.
I anticipated this from the beginning. You then, through the year, added the ASF and Spring planting US season weather factors to the continuing tariffs impact. All these factors are continuing, and yet you considered your July 15, 2018 point estimate as accurate. It wasn’t.
Providing unconditional point estimates without uncertainty is a bad practice when PREDICTING when public policy and public investment is involved. If that is the standard practice for economists, then it is a bad practice, which should be changed. At least some rational climatologists have moved away from this practice.
CoRev lies “in my experience of the Climate field”
CoRev, you HAVE NO EXPERIENCE IN THE CLIMATE field. Stop lying.
corev believes searching the web counts as experience in the field.
Dave, “in my experience of the Climate field” and you HAVE NO EXPERIENCE IN THE CLIMATE field.” Not what I said. If you want clarification: It’s my experience these types of predictions are discouraged or expected to include error bars or some uncertainty measures in the climate field.
Now stop lying about what was said. You are starting come across as a pgl clone.
CoRev can’t keep his facts straight again when he nonsensically emotes:
“Dave, “in my experience of the Climate field” and you HAVE NO EXPERIENCE IN THE CLIMATE field.” Not what I said. If you want clarification: It’s my experience these types of predictions are discouraged or expected to include error bars or some uncertainty measures in the climate field.
Now stop lying about what was said. You are starting come across as a pgl clone.”
Project much? You are such a joke. You literally have no experience “of” the climate field. Happy now?
“I am neither a statistician nor an economist.”
That is clearly the case. The 1st honest statement you may have ever made. Stephen Moore isn’t much of a statisticians or economist but he plays one on TV. Maybe we can get you to appear on the TV – Fox and Friends takes Princeton Stevie and he is an utter idiot. They should give you a call!
Hey, pgl, although he later said it was a “satire,” the only “information” CoRev has provided regarding his educational background is an apparently undergrad degree in women’s studies, this precious diploma hidden in a precious tube in a drawer his grandson cannot see, which also includes that medallion that was supposedly riding to the moon a half century ago at this moment with Armstrong, Aldrin, and Collins, a claim that I utterly do not believe and totally disrespect.
Sorry, CoRev, your claim that we are both “vets” of that mission I do not accept. Your claims and multiple lies related to this matter are at this particular moment for me who can really call it on you, absolutely and utterly beneath contempt.
Barkley, jealous? Don’t forget multiple diplomas and awards from the Apollo program (several from Apollo 11 and other missions). If you are going to mock, mock accurately. Otherwise you just look foolish.
Liberals do amaze.
CoRev,
Are we doing satire here again? “multiple diplomas”? On another thread you accused me of “lying” when I put “diploma(s)” empahasizing then how you only have/had one.
And “awards from the Apollo program (several from Apollo 11 and other missions)” Really? What are they? Again at one point you referred to a single “Apollo 11 award” but then later it became “awards,” although we have never heard what they were. You have also at time mentioned a “medallion” supposedly awarded to you and carried to the moon and back on Apollo 11. But when challenged on that you provided a link to medallions anybody can buy online now issued long after Apollo 11.
How stupid to you think we are?
And, btw, I went and checked again. There were awards given for the movie made about Apollo 11. Digging much deeper I did find that there were some “Apollo 1 awards.” However, there is no way you got one unless you are Buzz Aldrin. They were awarded by Copngress to the Apollo 11 astronauts and also John Glenn. Or maybe you are Michael Collins? Is he still alive?
Really, CoRev, 50 years ago at this time Armstrong, Aldrin, and Collins were on their way to the moon. I am not jealous about your clearly false claims. I am disgusted that you are persisting in lying about this matter at this time, although, again, perhaps you think you are engaging in witty satire. Sorry, it is not remotely funny.
It was 2011 when Congress made those awards to the astronauts. There is a gold medallion as well, which was made with the images of Armstrong, Aldrin, and Collins. It is in the Air and Space Museum, not in a drawer in your house with however many diplomas you have.
Oh, and CoRev, what on earth(or in space) does any of this have to do with political views. I make truthful statements about the space program because I am supposedly a “liberal who does amaze”? What?
Is the fact that you are blatantly making up lies about this due to you being a conservative? Do conservatives lie more than liberals because they watch fake news being spouted on Fox News all the time?
BTW, my late father, who did get awards that are a matter of public record, was a political conservative. However, he was also a man of honor and principle, so I suspect that if he were alive today he would be one of those Never Trump Republicans disgusted by him.
i would like to know exactly what expertise corev could possibly bring to the table to enable an apollo mission to succeed? based on his incompetence, at best, i would say he was responsible for the apollo 1 mission. so what, exactly, did you provide of such value to the apollo mission corev?
Barkley, you finally asked a meaningful question: “How stupid to you think we are?” Very, very! The rest of your comment is evidence of it.
Barkley Rosser Just a wild guess, but I suspect that when CoRev refers to an award or a medallion, what he means is something called a “Commander’s Coin”. Those coins are handed out to both civilians and military folks who were involved (sometimes only tangentially) in some project. The coins are handed out like candy, so I wouldn’t put too much stock in those kinds of things. Oftentimes the coins are used as a kind of team building tool. I usually pitch them, but I do have one kind of interesting coin that I use as a paperweight. Then again, not only am I arrogant (just ask CoRev), I’m also quite cynical.
2slugs, you’re much closer than Barkley. BTW, which version of his story do you believe? I think his latest, up on AB is probably the closest to the true impact of his father on the manned space program. He forgets that the BS flag was thrown because of his exaggerated version, and not that his father had an impact.
For those who do not go to AB, it is again slightly different from his ending version here, which was wildly different from claiming his saving the entire manned space program by “sending our astronauts into deep space.” https://angrybearblog.com/2019/07/a-half-century-since-apollo-11-launched-to-the-moon.html
If you want to compare story versions, I have the original version on another computer and if you want to see it, even again, just ask.
Baffled asks: “i would like to know exactly what expertise corev could possibly bring to the table to enable an apollo mission to succeed?” I was part of the tracking and data acquisition team which supported some late Gemini, Apollo, Skylab and several orbital and deep space missions of that time period.
A couple of these missions included lesser know successes, reviewing and tracing the AN/SRN-9 satellite data, a precursor to today’s GPS system (look it up). I don’t remember the timing of this associated with the Apollo missions, before 8, during or just after 17 .I also converted our star tracker program which drove the C and S band radar antennas to track the Sun to study the satellite signal strength needed to burn through the Sun’s background noise when we placed satellites in the Lagrange points between Earth and the Sun. Some of my other efforts were classified. All of these were team efforts of which I was just a member doing my part.
I doubt Barkley will understand these efforts, and I’m absolutely positive you are incapable of accepting these accomplishments.
Barkley, I always liked this Apollo 8 emblem. https://www.nasa.gov/sites/default/files/styles/side_image/public/thumbnails/image/s68-51093.jpg?itok=O9YXNYwF I have been sorry that I didn’t keep the awards and other memorabilia from it. I was too busy at that point finishing my education, raising a family and making a living, so treated them like I later treated that degree, thrown in the back of a drawer, and seldom remembered after receipt. It is only now, as an elder that I wish I had kept them. Barkley I find your lack of acceptance of my awards more an indication of your rocky relationship with your father, and carrying over those relationship impacts to others ” absolutely and utterly beneath contempt”.
Baffled and Barkley, I don’t know nor even care if you were/are part of such break through efforts. If so good on’ya! Barkley, I’ve said this before. You have every right to be proud of your father. His reputation doesn’t need your burnishing. Baffled, your inability to accept others experience, successes and anger that they even have them is a tell tale of your insecurities. I have probably processed more raw data in real time processing than you have ever seen. And, that was on fixed point computers, which were less powerful than the 1st floating calculator my wife got for me as a gift, WAY BACK THEN.
Both of you get over it. Others have successful experience in fields others than yours. Oh, and we’re proud of that experience and the associated awards. 🙂
CoRev,
Sorry, I do not believe anything you have written here, although maybe some of it is true. But you cannot keep your stories straight and what you wrote here contradicts things you have written elsewhere. The one time you previously said anything about your supposed role in the US space program and why you supposedly received a medallion or award, it was supposedly about doing something that kept costs down. Now we have a more dramatic account of being part of a team involved with important satellite tracking data and related matters. Also now we are told you did not keep any of your awards, but previously have told us that you had a wall on which you hung at least one of them so your grandson could see it, while the medallion was stashed in a drawer with that tube with the diploma (or is it “diplomas” again?)). Needless to say, none of this adds up at all.
Then we have the absurdity of you linking to my post on Econospeak from July 16 that was linked to at Angry Bear, with you claiming that somehow it contradicts what I have said about my late father’s role in all this. What a pile of crap, totally classic lying you, CoRev. I simply said nothing about my father’s role in these matters in that post. How can that contradict what I said elsewhere about what he did. As it is, you continue to claim repeatedly, as you have done in the past, that I have exaggerated his role, even as you now in the face of just too many links to public information about his career recognize that he was involved and did worthy things.
So, here on the 50th anniversary of the actual landing, I shall for the last time repeat my account of what he did, which has never changed, despite CoRev’s efforts to claim I did change it or that it is false or exaggerated, with another obvious lie by him on all this being his claim that he could “personally assure” us that the clocks were always synchronized throughout the Mercury, Gemini, and Apollo programs from the very beginning. So, the problem he was responsible for figuring out thus leading to it being fixed was that early in the program, well before Apollo, the ground clocks and ones on capsules in space were on different time systems, the ground ones on solar time and the ones on space on sidereal time. This meant that even when they started out synchronized at launch they would gradually diverge by about 1/365. This led to capsules landing further from where they were supposed to the longer they were up, with Glenn’s off by 40 miles. This got fixed. And, again, if it had not been fixed, then any attempts to go to the moon would have failed as they would have gone into deep space instead. I have added just a bit here, but this is and has been the story all along and completely correct without any exaggeration.
And I am still waiting for you to apologize for lying about this, CoRev. When will you? Today would be a very appropriate time for you to do it.
Oh, and since probably CoRev will probably put more bs up about this that further contradicts other things he has already said, I am going to try not to respond any further on this matter, which he has dragged out way too far already. I would suggest others try to do the same, although I fully understand how frustrating it is when he shows up with yet a fresh lie and distortion about something you have written here.
“I have probably processed more raw data in real time processing than you have ever seen.”
corev, considering the time frame you claim to have done your work, i would say this assertion is patently false. i can collect more raw data in one hour, today, than was captured in probably all of the apollo launches back in the day. the data throughput and processing do not even compare. i can capture terabytes of data in hours. you could probably fit all of the apollo data onto a large portable hard drive today. your have a complete ignorance of data collection, storage and analysis. i doubt you were given any work of significance during your “apollo” days. you certainly are not qualified to conduct any work of significance today. corev, being a nasa student intern a few decades ago does not make you a national hero today. we never would have gotten to the moon if people like yourself were responsible for the task in any way. it is clear to me that you are overextending your impact on those programs.
Barkley AMAZINGLY now gives a new and enhanced version of his original story. You’ve again made false claims of what I’ve said in the past because you were too lazy to go back and find the quote. Instead you relied on your faulty memory, which was influenced by misreading and misunderstanding what I wrote.
For an example of you poor reading and understanding, I said this: ” Barkley, I always liked this Apollo 8 emblem. https://www.nasa.gov/sites/default/files/styles/side_image/public/thumbnails/image/s68-51093.jpg?itok=O9YXNYwF I have been sorry that I didn’t keep the awards and other memorabilia from it….”
Your poorly functioning mind translated my comment to your latest lie: “Also now we are told you did not keep any of your awards, but previously have told us that you had a wall on which you hung at least one of them so your grandson could see it, while the medallion was stashed in a drawer with that tube with the diploma (or is it “diplomas” again?)). Needless to say, none of this adds up at all.”
That was Apollo 8 awards and not ALL my awards.
So let’s get to your original version of your story:
“So, I am going to go out on this by telling a real story, as I am completely disgusted by CoRev’s apparent claim that his somehow having some input into modeling a cost projection for some input to the moon mission was equivalent to actually physically modeling the moon mission itself. I mean, the lying jerk even capitalized his silly claim. So now I shall tell a real story about how the US got to the moon.
This involves my late old man, yes, J. Barkley Rosser [Sr.], whom I have always recognized as being more mathematically brilliant than I am and more important in world history than I am. This story is one reason why. My old man played a crucial role in getting us to the moon.
So, he had long been on the secret space committee of the US government from WW II on, one of its deepest advisers. As the Mercury, Gemini, and Apollo programs proceeded there was a growing problem. The longer they were in space, the further they landed from where they were supposed to. Yeah, CoRev, this is the actual physical modeling, and Houston was messing up. It was my old man who figured it out and then they got it fixed.
The problem was that the clocks on the ground and the clocks in the space vehicles were on different times, off by about 1/365. The ground was on solar time while the space vehicles were on sidereal time, tied to the stars. These systems were off by one day per year due to the revolution of the earth around the sun. So, the longer they were up, the further off they landed. Needless to say, if this had not been fixed, a shot to the moon would have missed with our astronauts going off into deep space to die.”
Later versions de-emphasized the Mercury Gemini missions. You’ve changed the time frame and defined it as “So, the problem he was responsible for figuring out thus leading to it being fixed was that early in the program, well before Apollo,…” You repeated your absurd exaggeration – if this had not been fixed, a shot to the moon would have missed with our astronauts going off into deep space to die, after saying it was fixed well before Apollo,…” Uh huh! A problem fixed for years could have sent our astronauts into deep space. Oh, you meant if not fixed, but it was. It would have been found in the early orbits because as we extended orbits the time drift would have been obvious and corrected as needed. You have forgotten the constant two-way data links I previously mentioned?
Your disbelief of my background is a serious detriment to talking about Apollo tracking.
Which part of my story do you not believe? Careful, you must quote it to prove you even understand what was written.
You sure are Baffled, makes a case for hows greatly technology has advanced in 50 years. He still can’t read and understand what is written. “I have probably processed more raw data in real time processing than you have ever seen.” I might add on a 30 bit, fixed point machine with a whole 32K 30 bit words.
BTW, if you want to compare history I ended my career working for DOD=wide in Command and Control and Combat Support systems. And you’re trying to impress with with data sizes and processing power? Please look these thing up instead of embarrassing your self even again: DOD-wide Command and Control and Combat Support systems. Here’s a hint 2slugs DLA, was just one feeder to the Combat Support systems, as were all the Services and other Agencies in DOD
YUP!!!! You really impressed me with your citing the difference in technologies. Oh, I’m sorry. You are clueless how far we old timers have helped improve technology. YOU”RE WELCOME! Now go wipe your snotty nose.
WOW! You can surely impress and amaze!
CoRev Here’s a hint 2slugs DLA, was just one feeder to the Combat Support systems, as were all the Services and other Agencies in DOD
I think something might have gotten garbled here. FWIW, I did not work for DLA, although I oftentimes worked with DLA on various projects. When it came to the occasional logistics project my counterparts at DLA were the folks at the old DORA office.
“I have probably processed more raw data in real time processing than you have ever seen.” I might add on a 30 bit, fixed point machine with a whole 32K 30 bit words.
corev, your comments simply reinforce my position. technology does make a difference. between you and i, i can guarantee that i have processed more raw data in real time than you have. this really has nothing to do with longevity, it has everything to do with the date it occurred. todays technology allows for far more data streams to be transmitted and processed in real time than ever before. you may have thought your work was large at the time, but history has shown past data capabilities are dwarfed every few years. your inability to even comprehend this aspect of data shows your ignorance of the field.
as an aside, your claims of past government work is simply evidence that government worker protections were simply too great in the past. somebody with your level of incompetence should have been released, but worker protections kept you on the taxpayer roles for far too long. what a pity. i am just thankful you were never tasked with anything of importance through the years, as lives most certainly would have been lost had you been given any real responsibilities.
Baffled, for you surely are, your failure to understand the magnitude differences of Command and Control and combat Support Systems versus ANYTHING you have seen in civilian organizations is breath taking, but indicative of your infantile knowledge. Riding on the accomplishments of folks like me, your you are impressed with yourself and the tools we in DOD developed or help develop. GPS and the internet are just a couple of the commonly known of them. OR DID YOU FORGET?
BTW, the first stand alone computer on a board was developed for Apollo, not just Tang. Your technological history ignorance is just breath taking.
Anything you’ve seen in the private sector is dwarfed in both size and complexity by its equivalent in one of the Military Services, and buried by the integrated DOD-wide systems. Your arrogance is due to your ignorance. What you forget, or more likely never knew, the size of these technologies you tout are there to support the very large users, such as DOD, and not the sum of you smaller folks. You ride on our accomplishments.
Why not give us a list of your accomplishments, and show how they were done without those accomplishments of your senior generation. I’ve shown just a few of mine. Show us which new technologies and how you helped implement them in your own illustrious career. I can. Can you?
Please wipe your snotty nose, and thank those of us who came before you. Without us you wouldn’t have the technologies you so love today.
i owe corev an apology. i never knew he was responsible for the internet, gps, computer boards, lasers and other advanced technologies. i assumed the world would be no different if corev never existed. boy was i wrong. you deserve a medal for all of these accomplishments corev.
but i digress. notice how corev has changed the argument from my questioning his contributions, to my questioning the contributions of the entire dod enterprise? now i will slap you on the wrist corev, and tell you to quit the grandizing. if you never existed, corev, we would still have those technologies-probably faster. i am fully aware of dod impacts. i have done quite a bit of r&d over the years with the military, and worked first hand on the dual mandates they require. but this is a side issue. my commentary to you has been quite direct and specific. i do not believe you have made any technical contributions over your career. why? because you have demonstrated you do not have any technical aptitude while commenting on this blog.
but corev, please keep on posting about how you led the charge in lasers, gps, satellites, apollo missions and the rest of modern technologies. you seem to have an imagination and ego that even dwarfs your dear leader trump. fascinating. in addition to your expertise in agriculture, climate, economics and the military, you are a technology savant. who would have thunk it! idiot.
Neither a statistician nor an economist, CoRev? Again, who knew?
Julian writes
“Again, CoRev is arguing for a specific type of trend analysis. This would generate the coefficients for a and b, which could be examined. If I understand CoRev properly, his trend analysis for labor is deterministic, and is estimated in a very similar way to the model above.”
With all due respect, this is utter nonsense. CoRev has not argued for any specific type of trend analysis. He has no understanding regarding coefficients for a and b. He doesn’t even understand the discussion about “FUTURES” prices versus “spot” prices, he merely parrots these words in cargo cult fashion. His entire goal is to obfuscate the discussion and spread doubt about whether or not Trump’s idiotic “zero-sum” approach to world trade and “negotiating” with China has any hope of success and the dangers involved with respect to penalizing domestic (US) producers AND consumers.
Well – CoRev does do trend lines in the same fashion that Lawrence Kudlow does trend lines. Dumber than rocks but serves the Republican agenda! And Kudlow heads the National Economic Council under Trump!
Dear Dave,
CoRev can make his own arguments about this. When he writes, ” I averaged the May/June average 2017 and 2018 prices. I was looking for a difference to estimate the 2018 early harvest prices assuming the trade talks had ended.”, that looks like a model to me. So I think AS is correct, that CoRev has a specific model that he or she prefers. I don’t agree with it, but it looks like a model to me. If I am wrong, as usual, then I am wrong.
Julian
Julian writes:
“CoRev can make his own arguments about this. When he writes, ” I averaged the May/June average 2017 and 2018 prices. I was looking for a difference to estimate the 2018 early harvest prices assuming the trade talks had ended.”, that looks like a model to me. So I think AS is correct, that CoRev has a specific model that he or she prefers. I don’t agree with it, but it looks like a model to me. If I am wrong, as usual, then I am wrong.”
And what are the coefficients for a and b in CoRev’s model? Oh that ‘s right there are none. CoRev’s “model” is that “May/June average 2017 prices” are more predictive of 2018 early harvest prices than soybean “FUTURES” prices. That is, the trade war would have no effect, because in his “model” the trade war was going to be a “blip.” What a joke. That is not so much a model as a willfull ignorance of what futures prices are and why they exist.
Julian, as we both will be.
Julian, let me add, we have again started comparing models versus the predictions derived from them. I still don’t know how to compare a conditional forecast versus an unconditional PREDICTION. Menzie has tried twice and I once, and we disagree on each others assumptions when trying to quantify.
I think we all agree that tariffs have lowered prices, I think Steven Kopits estimated in the10% range, but we have had little discussion beyond generalities of the two other major events, the ASF infection and US Spring Planting season weather. These also need to be considered with a set of secondary impacts, stored amounts, storage costs, degradation of stored beans, interest rates, etc. Some of these things are repetitive lending themselves to relatively simple estimation models, and are factored into carry costs.
The tariff effects are important but not the only price factors. On any given day can we predict which is has the largest impact? Today is the tariff or ASF impact dominating demand?
CoRev: Pretty sure with application of Bayes’ Theorem. But you never gave a probability of success in negotiations, nor did you at what horizon. So critical pieces of information are missing from you.
Menzie, you are absolutely correct: “you never gave a probability of success in negotiations…” Wat part of I was wrong is confusing you?
To what does this refer? “Pretty sure with application of Bayes’ Theorem.”
CoRev: One could calculate the probability of outcomes (and hence mean values) by using:
Prob(A|B) = [Prob(B|A)Prob(A)]/Prob(B) … You gave a conditional probability; I gave an unconditional. You can calculate the unconditional corresponding to your view as long as you know Prob(B|A), Prob(B).
Or, that’s what I learned in high school.
Menzie admits: “I gave an unconditional.” and then further confuses with It’s so simple to calculate the probabilities, he learned it in HS.
Let me repeat. It was a bad practice to ignore such a simple calculation in your unconditional point PREDICTION. It established conditions that were almost impossible to meet. Worse, you consider that standard practice in economic estimates/forecasts/predictions upon which decisions affecting huge investments can be made.
At least rational climatologists have moved from such a bad practice because of challenges to its use from within and without the field.
CoRev: You have now completely convinced me: you are an utter idiot, and worse yet, unwilling to admit when you are in a completely indefensible stance. If I get to teach stats again, you will be a canonical case for statistical illiteracy.
Menzie, I truly hope you do, but when you please quote my comments comparing the differences between climate and economic field estimating practices and why. Failure to do so will just be continuing a teaching disservice to your students.
Yes, I feel as strongly as you about this estimating practice. You now know my motivation for wishing this standard practice to change. What’s your for continuing it? Because that’s the way we were taught and we all do it?
You’ve lost all objectivity. It doesn’t take much to convince in that frame of mind.
CoRev: No, I do things the way I do (which does not always conform to practice) because I think they make sense. Otherwise, I’d be using Rat-Ex like I was taught, etc.
CoRev is basically in agreement with Trump on models: very little (or no) color, more curves, less reliance on older ones. Foreign models are acceptable as long as their countries of origin are not _______holes.
CoRev,
Maybe I have not read your desired forecasting method closely enough to understand the method. It would be helpful if you could succinctly state the principles of your forecasting method.
Regarding Professor Chinn’s knowledge and ability, surely you cannot believe or think that others will believe that Professor Chinn does not understand regression modeling, statistics and the various characteristics of valid regression models. From the multiple postings of his regression models, notice that he always indicates the statistical significance of the regression variables. Often, when he posts charts with a mean estimate, he indicates the one or two standard deviation bands. Looking at his academic papers he posts, notice that he shows various measures of model accuracy such as MSE.
The forecast of the soybean July 2019 closing price was a special case forecast as he has indicated which relied upon the price of the futures one year prior. I don’t understand the carping about the soybean forecast and the subsequent generalization about Professor Chinn’s skills as an economist, instructor and as a forecaster. What seems like food fight arguments undermines any valid argument that you may have regarding forecasting or modeling.
AS, the best I can do is ask you go bakc and read my July 20 0601 and 0620 comments to Menzie. That give my reasoning why I am concerned.
You misunderstand my questions of Menzie. They are not aimed at his experience, his model, not his knowledge, but ARE AIMED AT HIS SPECIFIC WORDING OF HIS July 15, 2018 PREDICTION.
“Conclusion
Soybean futures are remarkably good predictors of future spot prices of soybeans. Hence, my best guess of soybean prices one year from today is 872. ”
Compare those two points, July 2018 and 872, against the July 15 2019 and $9.02. Those were the only data included in his “Conclusions statement”. His statement was incomplete making his PREDICTION incomplete/inaccurate.
Why I emphasize this important point is contained in those two comments I referenced .
CoRev,
So if Professor Chinn, had stated his forecast as $8.72 + – $0.40 at the 90% confidence level, the argument would be over concerning the presentation of his forecast?
AS: Since I’ve never seen CoRev provide a confidence interval of that nature (e.g., “90% C.I” let alone +/0 2 standard errors), I doubt it…
AS, yes, absolutely! Like your interest in economics, I have had a similar in Climate Science for ~ 20 years.
Another point everyone has missed and forgotten was that I was willing to give Menzie the benefit of the doubt. I had already given my approach to validating his PREDICTION, but his incessant mocking forced me to re-think that proposal. I don’t remember seeing any MODEL validation method in any of his posts, so I provided my own. I do remember thinking after his now infamous Conclusion statement, how did he validate the model to be so confidant to get that unconditional result?
This is what I had previously proposed:
”
CoRev
May 13, 2019 at 6:10 am
Using today’s 1 year/52 week prices, we find that US soybeans spot prices traded with a +/- ~12.5% range, $10.51 to $7.87. Since Menzie’s $8.72 prediction soybeans spot prices traded in a much narrower band, +/- ~3.6%, $8.14 to $7.87.
The question is will Menzie’s $8.72 prediction fall within the trading range since it was provided, 7/15/2018? The test is whether on 7/15/2019 spot prices fall between $9.04 or $8.4, +/- ~$.32? If measured today, it is a fail.”
It kinda gives a lie to Menzie’s claim below: “AS: Since I’ve never seen CoRev provide a confidence interval of that nature (e.g., “90% C.I” let alone +/0 2 standard errors), I doubt it”
CoRev: A trading range is not a typical statistically defined prediction interval. That’s a concept from the world of technical analysis…
Menzie, and you expected a typical statistically defined prediction interval from a non-statistician? For a year i was waiting for you to handle the obvious uncertainty issue, and you still can’t admit it was a problem.
How did you validate this model’s results?
CoRev: The model was validated using out of sample forecasting, in Chinn and Coibion (2014). Do you still not understand that? Anybody who has done an elementary horse-race comparison of forecasts understands.
Menzie,if I’ve got this correct, neither the values and the conclusions show in both the paper and the article indicate the certitude shown by the 872 unconditional unqualified conclusion. IIRC, Chinn and Coibion, 2014 and the article claimed the methodology better than the others studied while the details included never showed the certitude of your “best guess”.
How could you have made the unqualified 872 PREDICTION? Only as we approached the selected date did you revise your predicted price, but still refrained from showing the associated uncertainty.
Because that’s how we’ve always done it, is a poor argument for continuing the practice. If someone, who most here consider stoopid, knew it was overwhelmingly likely to be wrong, as it eventually was, why would you continue or have you continued here for a year?
If you consider this a minor or nonexistent problem, then we very much disagree. As will the vast majority of the “stoopid” populace.
2slugs, AS and Julian EM is starting his consolidation of findings and “how to” of his Global Historical Climatology Network. Here: https://chiefio.wordpress.com/2019/07/21/ghcn-v3-3-vs-v4-top-level-entry-point/ If you have any interest in his current results and analysis.
Since you have worked with Dr Spencer and some of the temperature data this would be a chance for you to easily download and format into a database for your independent study. You might even be able improve on the climate statistics standard practices.
2slugs, AS and Julian if you don’t want to go be troubled OK, but you might just follow along with his efforts to explain the hill of sand upon which climate is built, and why we skeptics don’t believe climate change catastrophe predictions.