From 2019, pay special attention to CoRev’s statistical analysis (at the end):
Reader JBH writes:
“Employment has done marvelously well under this president.”
I laughed and laughed and laughed when I read this. Why? Take a look:
Figure 1: Nonfarm payroll employment (dark blue), and stochastic trend (red). Stochastic trend estimated using 2010-2016 data, and regression of first log difference on a constant. Source: BLS March 2019 employment situation release, and author’s calculations.
We are doing as well as we were from 2010-2016, even after massive tax cuts and the end of spending restraints amounting in the trillions.
But kudos to JBH for livin’ in a fantasy world … it for sure is more pleasant than my world. Thanks, Drumpf!
Update, 4/12/2019, 4:15PM Pacific: Reader CoRev requests:
Can you provide the raw data used?
Here is the dataset used, directly downloaded from FRED.
Update, 4/16/2019, 9AM Pacific: Reader CoRev has provided his check-file on my trend analysis. Without comment, here it is:
Some three years after CoRev provided his “anomaly analysis”, I still don’t fully understand what he’s doing. It looks like deviations from averages. If anybody can tell me why it validates his view of the world (i.e., my choice of use of stochastic trend and time sample biases against seeing a boom in Trump employment up to that point), please tell me. I am (still) dying to understand.
In the meantime, here is a graph demonstrating the trouble with linear (deterministic) trends as applied to nonfarm payroll employment (reprised from “Why Friends Don’t Let Friends Apply Deterministic Time Trends to Nonfarm Payroll Employment”):
Figure 1: Nonfarm payroll employment, 000’s, s.a. (black), and linear deterministic trends estimated over 20 year subsamples. NBER defined recession dates shaded gray. Source: BLS, May employment situation release, NBER, and author’s calculations.
With a log scale? Maybe someone finds that easy to interpret, but it’s pretty much gibberish to me, graphically speaking.
Steven Kopits: Yes, I know it’s gibberish to you.
Anyone who studies macroeconomics for periods longer than say your one-week attention span gets the need to show growing variables using a log scale. In fact a lot of the math for growth models is done in natural logs. Oh wait – I said math. Already over your limited capabilities. So do us a favor – go bug some other blog with your whining.
Ah yes, natural logs. Brings back not so fond memories of when I took quantitative chemical analysis in the pre-calculator days. We had to use 5 place log tables to do all the calculations on triplicate samples. Kids to day have it so easy; they don’t even have to walk 3 miles through the snow to school any more.
Up hill. Both ways.
Let’s pretend for a few moments that “Princeton”Kopits genuinely wants to know:
https://blog.supplysideliberal.com/post/31224211784/the-logarithmic-harmony-of-percent-changes-and
And if he doesn’t, “other students in the classroom” may learn something.
Steven,
Maybe the few comments below can reduce the feeling of gibberish.
Looking at the payroll log scale presented, notice that each vertical scale numeric change is 20,000k, but the percentage change progressions represented by each horizontal 20,000 progression show a smaller percentage change as shown below and on the presented graph by the vertical distance between y-scale values.
Notice that the vertical distance for the payroll change from 80,000k to 100,000k is half the distance of the change from 40,000k to 60,000k to account for a 25% change and 50% change respectively.
Employment in Thousands below:
160,000 percent change from 140,000 is14%
140,000 percent change from 120,000 is17%
120,000 percent change from 100,000 is20%
100,000 percent change from 80,000 is 25%
80,000 percent change from 60,000 is 33%
60,000 percent change from 40,000 is 50%
If a chart for the Shiller stock market data from 1871 to 2021 were presented in linear and log scale, it would be more dramatic. The linear scale would look like a hockey stick, whereas the log scale or the use of log values would show percentage changes, which is what I think most of us consider when thinking about stock market changes. Even Shiller’s linear scale real S&P 500 composite index chart shows quite the hockey stick. See the “Index Plot” tab.
https://view.officeapps.live.com/op/view.aspx?src=http%3A%2F%2Fwww.econ.yale.edu%2F~shiller%2Fdata%2Fie_data.xls&wdOrigin=BROWSELINK
Oh dear, Steven. This is pretty basic stuff, I mean like high school if not junior or middle school level stuff. I do not think you want to announce to any of your potential clients that you made such a comment here or anywhere. I suggest you “suppress” the knowledge that you did so.. This is the sort of thing that could make the likes of CoRev look almost on top of things.
Stevie linked to something from the FED that was a high school presentation of macroeconomics declaring it to be some worthy macroeconomic model. Seriously – he is THAT dumb.
I don’t think that graph is showing what it purports to show.
Menzie writes: “here is a graph demonstrating the trouble with linear (deterministic) trends as applied to nonfarm payroll employment (reprised from “Why Friends Don’t Let Friends Apply Deterministic Time Trends to Nonfarm Payroll Employment”)…”
It’s not a linear graph. It’s an log graph. If you want to show linear problems, you use a linear scale and show that linear estimations do not fit the data. Btw, you can make the data fit a very nice linear approximation on a log scale all the way from 1947 through 2000. After 2000, everything falls apart.
As for linear approximations, I think they can work fine over shorter, say ten year periods, just as Fig. 1 shows.
To the broader point, hiring appears to have its own speed limits largely independent of policy, an interesting topic in its own right.
And to think, I was building up so much anticipation CoRev was going to get the John Bates Clark Medal in 2023. Another setback.
I think you are pointing to two important issues when evaluating how well things are going in a specific period of time. One is whether things are simply continuing on the previous/natural trajectory. If it does you can say that the President at least didn’t screw up what the previous President had put in place. The other issue is specific for economic parameters; did this (positive effect/ apparent lack of screwing up) result from a huge increase in national debt. It is easy to be/act rich if you first take out a huge loan.
A couple of posts ago, when CoRev was tossing as much dust in the air about recent climate research (which he insisted was different from weather research because of some nonsense he spewed) as he could, he finally got around to claiming that Menzie necessarily misunderstanding climate research because he’s an economist, and nobody (except, apparently, CoRev) can understand two fields of study.
Of course, the implicit assertion behind everything CoRev writes is that he understands stuff. Lots of stuff. Like economics (which he doesn’t understand) and climatology (which he doesn’t understand) and geopolitics (nope)…..
There’s a commonality between Corey’s pose about his own knowledge and his attempts to deny other folks’ knowledge – Corey’s pose is meant to deceive and his attempts to deny other people’s knowledge is also meant to deceive.
CoRev is a bought-and-paid-for tool. Menzie is a well respected expert. It’s a shame Menzie’s time must be spent correcting Corey’s lies, but that’s the price of a liberal comments policy.
I am going to take the opportunity to follow up on macroduck’s comments, which I totally agree with aside from doubting that CoRev is bought and paid for (who would pay for such an incompetent idiot?), to poke further at CoRev but also a bit at Moses, whom macroduck seems to continue to like a lot.
So, in another thread Moses declared in two posts that i 1) “make up s**t,” and 2) move goal posts a lot. I responded to both of those charges by noting that I have never done either of those, although I do make mistakes from time to time, and admit them when they are pointed out.
However, I shall note that the person regularly commenting here who most regularly and frequently is guilty of both of these things Moses accused me of is indeed none other than our CoRev. And this last round on the matter of temperature shocks was simple a massive display of both, with macroduck laying out at least some of it. He started out with this irrelevant stuff about weather and climate, which he returned to at the end, but then after macroduck and AS and some of the rest of us pointed out that he seemed not to understand what was in the paper, he moved the goal posts and got tangled in a mess on how to measure trends. But as he got taken to task on that more, he suddenly shifted over to the economics side of it, dragging in an old paper by the climatologist Pielke that relied on an odd measure of sustainability to claim that economic costs of weather disasters were declining. Macroduck then pointed out that actually that paper showed them rising, but CoRev noted that they were declining as a share of GDP (which I think is not still true anymore). But then he reverted declaring either that indeed economic costs were declining or that they were “zero” in the end.
As it is, CoRev has on many occasions provided us with such spectacles of made up falsehoods (“s**t”) combined with constantly moving goal posts when people nail him on his falsehoods. “Oh, maybe I was not quite right about Issue A, but actually Issue B is what is really important, and let me introduce yet another false statement here, this one about Issue B!”
And, yes, his ongoing efforts to declare himself a great expert on many things, more than others here, despite his record of making one erroneous statement after another is really quite astonishing. In this one he did indeed declare himself to be both a better economist and climatologist than anybody here. Of course, he has been making such claims since he told us he was a better soybean farmer and expert on futures markets than anybody else, along with his famous assertions about his role in the US space program, the bottom line truth of which we never were able to ascertain after his many conflicting claims.
barky,
you and moses are such a treat!
i have discovered why you adore the helsinki accords as if they are biblical:
https://niccolo.substack.com/p/delusion
the osce is the usa’ prime functionary on the accords and their plan is portrayed!
i used to think it senility, now i suspect more.
Anonymous,
Oh, with this you are back to making me suspect that you actually are a Putin bot after all and not just a very dumb American. As it is, this is the first I have heard of this plan that the Helsinki Commission is apparently about to hold hearings on. It does seem to be rather over the top frankly.
Indeed I am a fan of the Helsinki Accords, without which I would not be married to my wife. Of course, this might please Moses as I would no longer be reporting here any of the sorts of inside information from Russia that I have done, although that would include my one blunder in the runup to the war criminal invasion of Ukraine by Putin, the moment when I agreed with Ukrainian President Zelenskyy to believe V.V. Putin when he spoke to the Russian people in Russian telling them that the troops he had sent to Belarus for exercises would be coming home after they were finished with their exercises. I quickly figured out he was lying the minute Belarusan President Lukashenka came out with his bizarre demand that the troops remain to protect Belarus against a possible invasion from Ukraine. But perhaps you think that latter was a serious danger. As it is, I have long granted Moses that I was mistaken to believe Putin’s speech making that promise that I reported here, obviously a sign of either senility or perhaps dead end neo-connism on my part in whatever you have for a mind.
Oh, I would note, although I think I have only mentioned it once or twice, that indeed Putin and Russia are clearly in full violation of the Helsinki Accords with this invasion of Ukraine. But then he and Russia are also in violation of the UN Charter, which Ukraine was actually a separate charter memver of, thanks to a demand by Stalin. As it is, I think here I more frequently noted that Putin and Russia are in violation of the more recent Budapest Memorandum of 1994 that involved Russia, along with the US and UK promising to respect the sovereignty and territorial integrity of Ukraine if it gave up its nuclear weapons, which it did. Do you wish to somehow argue that Russia is not in violation of any of these agreements, “Anonymous”?
Curiously, Putin has put sanctions on the members of this Nelsinki Commission. Well, I guess that will show them!
Gets pretty ugly on the “niccolo” web link there. Almost enough to make a sicko like me blush. [disparaging references to other commenter’s physical appearance will henceforth be deleted – MDC]
So, I just checke on this Niccolino Soldo more fully, as well as this post. First of all, it should be noted that that the claim that this group is calling for “partition” of Russia seems to be bogus. I grant that this talk of “decolonization” might well imply that possibility, but it could also imply increased autonomy for some of the ethnic minority republics and oblasts in Russia without actual partition.
As it is, this Soldo is really a piece of work. On marginal Revolution of all place he has been labeled a “far right neo-fascist,” and he seems to be obsessed with being “anti-woke.” His general discussion of the war situation is over the top Putin troll fascist and misinformed. That “Anonymous” thinks this is something impressive to present here says quite a bit, although probably not that he is an actual Russian. I am back to thinking he is just a very dumb American. After all, an actual Russian troll would have come up with some all-American sounding fake name like “Jack Smith” or whatever rather than coming on as “Anonymous,” which just makes him look as stupid as he seems o be.
All that said, I shall agree that it looks to me not to be all that wise in the current situation for this group to be holding such a session, even if Soldo has oversold what it is. It obviously provides fodder for the putin propagandists to do just what we have seen here and for no good reason as there is no way anybody is going to act on any of this, however justified it might be.
@ Menzie
Booooo!!! Hiss!!!!! Booooo!!!!! Hiss!!!! Perversion humor killer!!! And after all the societal progress me and Larry Flynt have made on this issue.
(just being silly here) My base (no pun intended) intention is never to sully your blog Menzie. I have enough problems not sullying myself). I owe you about 20 times (100 times??) over for saving me from myself in comments. I wouldn’t do that to you Menzie. It’s my badly needing anger management genes kicking in.
Barkley is kinda the Richard Cole to my Jimmy Page. He’s asked many times to carry my “equipment” around to the next music gig but I told him he’s too frail to lift heavy things. I just keep him around to scare off the ugly contingent of my groupies.
How many times are you going to re-hash this approach? IIRC this is your 2nd or even 3rd attempt to understand it. You now claim after 3 years:
“Some three years after CoRev provided his “anomaly analysis”, I still don’t fully understand what he’s doing.”
I don’t understand the confusion over this simple and common usage for climate temperature data. ttps://iridl.ldeo.columbia.edu/dochelp/QA/Basic/anomalies.html
How are anomalies calculated?
An anomaly is the difference between an actual value and some long-term average value.
For example, if
X = actual value of average temperature for January, 1982 and
Xbar = long-term average temperature for January (an average over many years)
anom = anomaly value for January, 1982
then
anom = X – Xbar
But you readily accept this anomaly-like approach in this article: https://econbrowser.com/archives/2022/06/guest-contribution-does-monetary-policy-respond-to-temperature-shocks#comment-277517
“We start by collecting average temperatures in each county at daily frequency. For each series, we group observations by quarter and compare the within-quarter distribution of temperatures to a reference distribution, which is made by pooling daily observations recorded in the same quarter of the past years. We argue that five years is a sufficiently long period for agents to figure out the shape of the underlying temperature distribution, so we construct the reference distribution based on that time span.9 The reference distribution rolls over time,i.e. it is updated every year for each quarter”
Incidentally the article’s source of data was the NRCC who shows its data source: “Sources: CISESS and NOAA NCEI. Data (a, b, c) GHCN-Daily from 10 (MD) and 655 (CONUS) long-term stations, (d) nClimDiv. ” Most of these data are anomalies, but Filippo did not give the link to his source dataset.
Oh, Joy! He’s at it again!
CoVid, you do realize that most of what you’ve written here gives the impression of meaning something, but actually doesn’t? I think you do. It would be really, really unlikely you could accidentally construct that much meaningless text which gives the superficial impression of meaning something. Not chimps typing Shakespeare unlikely, but close. Intentional obfuscation is a much simpler explanation.
I know, you’ll just say “No! Uh, uh! You guys just don’t comprehend my genius!” Kinda like what you just wrote to the PhD-toting, university-teaching, White-House-advising host of this party. (You realize I left some stuff out, right?) Very credible, by the way. Problem for you is, the only people here who aren’t on to you are the ones who behave just like you. That’s funny for us, but sad for you.
CoRev,
Population and thus labor force is growing at a percentage rate, which is appropriately estimated by using log differences as Menzie indicates. The amount of new labor is a pereentage of the amount of previous labor.
This is not appropriate to looking at temperature trends because temperature is not a population growing as a percentage of its recent levels. It is growing at a rate correlated with rate of accumulation of GHGs in the atmosphere, which are not generating themselves in a percent way as population does with itself, but as an accumulation of exogenously injected amounts of the GHGs. These are quite different processes. So, temperature may be rising linearly whereas population rises exponentially.
“temperature may be rising linearly whereas population rises exponentially.”
And we thought CoRev believed there was no long-term rise in temperature at all! After all – we do not use logs simply because a series has variability. Of course CoRev understates this issue less than even his feeble knowledge of long-term economic growth.
CoRev: Do note that the authors of this paper use a five year trailing moving average to define a “shock”.
Menzie now claims: ” Do note that the authors of this paper use a five year trailing moving average to define a “shock”.”
You do note that is the baseline for comparing: “We start by collecting average temperatures in each county at daily frequency… compare the within-quarter distribution of temperatures to a reference distribution, which is made by “pooling daily observations” recorded in the same quarter of the past years….We argue that five years is a sufficiently long period for agents to figure out the shape of the underlying temperature distribution, so we construct the reference distribution based on that time span. ”
Five years is the reference period to which the county DAILY temps are compared. They accumulate these hot/cold shock days by quarter. ” County surprises are the difference between the number of extremely hot and cold days in quarter t and the number of extreme days in the reference distribution for that quarter.”
All my quotes are from the paper with the exception of my quote of your comment.
The best wisdom I know of regarding the use of evidence to persuade those for whom “alternative facts” will do just fine comes from a 17th century English reform preacher:
“We mistake men’s diseases when we think there needeth nothing to cure them of their errors but the evidence of truth. Alas! there are many distempers of mind to be removed before they receive that evidence.”
— The Practical Works of Richard Baxter
CoRev’s mind is distempered, so that evidence of truth cannot reach him. How those distempers can be removed, I cannot say. Baxter’s advice was to speak to sinners privately, because men are reluctant to admit their sins in public. How that could work in our webby world to cure CoRev’s distempers, I cannot imagine.
MD, I see you don’t understand anomaly calculations either. You should, as you’ve written about temperatures often enough. You appear unaware that many of the various temperature data sets are in anomaly form. Are you actually that ignorant?
Wow – more weaving big words together without a shred of actual thought. You should write a book on how to excel at disinformation.
Barking Bierka – the NYC Jerk, if those are big words to you, then you need to ask for your diploma back. I can’t imagine you with a degree.
Hi Menzie
In another misunderstanding of logarithmic growth – ( https://en.wikipedia.org/wiki/Logarithmic_growth) I thought that more than a million excess deaths would convince Trump and the MAGA crowd of the reality of biological exponential growth – but – no: https://doggett.house.gov/media/blog-post/timeline-trumps-coronavirus-responses
Menzie claims: “Some three years after CoRev provided his “anomaly analysis”, I still don’t fully understand what he’s doing. It looks like deviations from averages If anybody can tell me why it validates his view of the world (i.e., my choice of use of stochastic trend and time sample biases against seeing a boom in Trump employment up to that point), please tell me. I am (still) dying to understand.”
Yet ~ 12 years ago Menzie provided this definition: “https://econbrowser.com/archives/2009/11/the_global_surf
“From Temperature Anomaly FAQs:
The term “temperature anomaly” means a departure from a reference value or long-term average. A positive anomaly indicates that the observed temperature was warmer than the reference value, while a negative anomaly indicates that the observed temperature was cooler than the reference value.
The reference value used to create this graph was the average over the 1901-2000 period.”
Does he still misunderstand the anomaly process and its use in determining temperature change? Or is its use in in another time series just not what he is confused with? Is it the best test? Dunno, that’s a value judgement. It is the approach that climatologists use though.
CoRev Menzie’s post questioned your not understanding the difference between a deterministic trend and a stochastic trend. Stochastic trends are not mean reverting and have unbounded variance. Stochastic trends are nothing more than the accumulation of random shocks that give the illusion of a trend. So if you have a large positive shock that is followed by a few smaller but random shocks, the accumulated effect will appear like a trend. A deterministic trend makes the change in “y” a function of time and is mean reverting with a constant variance. You so called “anomaly analysis” is little more than a ball of confusion. It probably makes sense to someone who doesn’t have a rigorous understanding time series analysis, but that says more about your education level than it does the actual data.
You should also pay attention to Barkley’s comment about the difference between an “anomaly” and a log difference. Log differences are used to make nonstationary stochastic trends stationary by subtracting out the previous period’s level. An anomaly is just the difference between an observed level and some longer run mean.
Climatologists use temperature anomalies to display graphs showing long run changes relative to some long run mean. Climatologists do not use anomalies to do time series analysis of climate change data. The kind of time series analysis (and emphasize analysis as opposed to creating a graph for public consumption) that climate scientists use is way over your head.
You are forever telling us that economists don’t have any special expertise in climate science. Aside from the fact that economists probably do have some special expertise in the time series aspect of climate science, I would agree. That’s why I accept the scientific consensus. I’ve read books and papers on climate science and I readily admit that I’m not an expert. I’m an informed amateur. But let’s be clear. You are not an expert either. You don’t have the math skills. You don’t have the most basic understanding of time series analysis. You are an uninformed amateur.
Let me conclude this comment with this question. If the anomaly process is not accurate for time series, then what makes it accurate and appropriate for the climate, in particular temperature? So which is more correct your hyperbolic disbelief of this approach for nonfarm payroll employment totals versus your unquestioning belief in temperature anomalies? You obviously believe the analyses from these temperature anomalies, since you advocate policies costing us $ trillions per year and to which many believe is a major cause for today’s world-wide inflation.
Think carefully as your hypocrisy is on the line.
CoRev Huh??? This is complete gibberish. What is wrong with your brain that you somehow think this is even intelligible enough to address? What you’re call an “anomaly process” makes for useful graphs and it shows that temperatures are increasing. But that is not anyone’s idea of “analysis” in the sense of time series analysis. For example, displaying anomalies tells you nothing about the stationarity of the time series. Menzie’s post referred to deterministic versus stochastic trends. Conducting tests to decide which whether a time series is deterministic or stochastic is (quite literally) the first step in any time series analysis. Your idiotic “anomaly analysis” completely ignores this. It isn’t a question of believing or not believing in temperature anomalies. Clearly you can make a graph displaying temperature anomalies, and those anomalies clearly show a rise in global temperatures. The interesting question is whether a trend is deterministic or stochastic. Unfortunately you wouldn’t know that because you clearly have no idea what those terms even mean.
2slugs, do you actually believe daily (or even more frequent) measurements of temperatures are deterministic? Why not take it up with climatologists who use frequently use it? They clearly don’t care whether a trend is deterministic or stochastic when they use anomalies.
To boil down your complaints, I applied a successful tool used by climatologists, and not used in economics. A tool where stochasticity, deterministicity and stationarity are not required to use that tool. For years you have blindly accepted the analysis of climatologists. Do you no longer so believe?
You also claim: “Conducting tests to decide which whether a time series is deterministic or stochastic is (quite literally) the first step in any time series analysis. ” Not true for all fields and all tools as I have shown.
Your whole complaint is hypocritically gibberish.
CoRev Not true for all fields and all tools as I have shown.
Excuse me, but you are the one who used something you called an “anomaly process” to examine “trends” in employment. The only thing you’ve shown us is that you have absolutely no understanding of time series analysis.
I applied a successful tool used by climatologists, and not used in economics. A tool where stochasticity, deterministicity and stationarity are not required to use that tool.
Then you haven’t read much climate science. Most grown-up climate studies do in fact ask if temperature data are stochastic or deterministic. Displaying anomalies is fine as far as it goes, but it doesn’t answer the more interesting questions.
Question: when we talk about stochastic trends and deterministic trends, do you even understand the difference and why it matters?
CoRev For you edification:
A paper arguing for a deterministic trend:
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0060017
And an argument for a stochastic trend:
https://scholar.harvard.edu/files/stock/files/doestemperaturecontainstochastic.pdf
Or this paper that essentially finds a cointegrating relationship across three stochastic trends:
https://www.osti.gov/etdeweb/biblio/20824005
…the results indicate that the time series for temperature, anthropogenic emissions of CO2 and CH4, and
their atmospheric concentrations contain a stochastic trend. Traditionally, analyses
of the temperature record avoid the assumption of stochastic trends because these
trends are characterized by their long-term memory – the effects of innovations
do not fade over time. As such, temperature would be inherently unstable with no
tendency to return to a long-run mean.
This seeming contradiction is reconciled by identifying the sources of the
stochastic trends. The stochastic trends in temperature are caused by stochastic
trends in the radiative forcings that drive temperature, and not temperature itself.
That is, a direct shock to temperature does not accumulate over time. Rather, the
stochastic trends in temperature reflect the stochastic trends in the radiative forcing
of greenhouse gases and anthropogenic sulfur emissions.
Let us know if you understand a single word of all this.
2slugs, more BS. You’re getting desperate. “To boil down your complaints, I applied a successful tool used by climatologists, and not used in economics. A tool where stochasticity, deterministicity and stationarity are not required to use that tool. For years you have blindly accepted the analysis of climatologists. Do you no longer so believe?”
Your whole complaint is hypocritically gibberish.
CoRev: You *do* know that if a series is I(1), then there is no population mean that you can estimate…
2slugs, did you note this from your reference? https://www.osti.gov/etdeweb/biblio/20824005
“We use recent advances in time series econometrics to estimate the relation among emissions of CO2 and CH4, the concentration of these gases, and global surface temperature. ” Are you as shocked as I that someone using econometric tools would do stochastic or deterministic tests.
And this: ” The regression results also indicate thatincreases in surface temperature since 1870 have changed the flow of carbon dioxide to and from the atmosphere in a way that increases its atmospheric concentration. ”
Or did you note this: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0060017
” using modern econometric techniques.” Are you as shocked as I that someone using econometric tools would do stochastic or deterministic tests.
Or that this paper https://scholar.harvard.edu/files/stock/files/doestemperaturecontainstochastic.pdf appears to be a defense of the 1st paper referenced above.
And why did you ignore this comment? “They clearly don’t care whether a trend is deterministic or stochastic when they use anomalies. “
Menzie,
It does not matter what CoRev knows. Once CoRev gets going with his goal post moving, it is all over, no matter how accurate or correct the person trying to argue with him is. Poor 2slug is doomed. I mean, CoRev has him! 2slug suggested that CoRev was speaking “gibberish,” well, now CoRev has twice replied that it is 2slug who is speaking “gibberish,” victory for CoRev! 2slug points out papers variously estimating deterministic and stochastic trends, including with such modifications as using cointegration, and, well, CoRev lets 2slug know that sometimes climatologists use econometric techniques! How can 2slug win? And then we had that glorious moment when CoRev really got to a new and important goal post by noting that 2slug supports spending trillions of dollars a year on all this! I mean we clearly saw him doing that, didn’t we???
Menzie, you do know we are talking about temperature change don’t you? It is a change we live with hourly/ daily/ seasonally/ annually/ and periodically. Do you believe it is deterministic? More importantly, if you run the 1st deterministic test do you believe it could change?
How many times does the test need to be run on temperature change, especially if you are doing anomalies?
Barkley moves the goal posts even again, but fails to answer the core question I’ve raised: “And why did you ignore this comment? “They clearly don’t care whether a trend is deterministic or stochastic when they use anomalies. “
If this is goal post moving then you haven’t read this and the other associated threads/articles: “To boil down your complaints, I applied a successful tool used by climatologists, and not used in economics. A tool where stochasticity, deterministicity and stationarity are not required to use that tool. For years you have blindly accepted the analysis of climatologists. Do you no longer so believe?” “To boil down your complaints, I applied a successful tool used by climatologists, and not used in economics. A tool where stochasticity, deterministicity and stationarity are not required to use that tool. For years you have blindly accepted the analysis of climatologists. Do you no longer so believe?”
You tout 2slugs where he found 3 papers done by those familiar econometric and climate and writing in both fields. Incidentally 2 of those 3 papers were by the same authors.
Did you actually read any of 2slugs papers or just blindly use your go to comment re: goal post moving?
You’ve been making egregious mistakes lately, and this is another.
That was a repeat of the point, and this time I even highlighted it just for you.
Just for Menzie, Barkley and 2slugs, the subject under discussion is the use of anomalies when showing world-wide average temperature changes. These data are averaged temporarily and spacially. Do they need a deterministic and/or stochastic test? What value would it/they provide? How often would it need to be run?
What amazed me is that none of you have commented on the 1st Kaufmann paper finding this: “The regression results also indicate that increases in surface temperature since 1870 have changed the flow of carbon dioxide to and from the atmosphere in a way that increases its atmospheric concentration. ” I even highlighted it.
Econometrics at its best?
CoRev: Actually, the post was about using deterministic trends when talking about a series — nonfarm payroll employment — that almost everybody agrees is I(1). That’s the series to which you applied your “anomaly analysis”.
Menzie, you do know that we are talking about arithmetic averages and not statistical means don’t you when we calculate anomalies?
You start off saying: “Some three years after CoRev provided his “anomaly analysis”, I still don’t fully understand what he’s doing.” After 3 years and at least 2 if not 3 attempts to ridicule it or me, you still don’t understand the anomaly process, but still make value judgements. Worse this process is used in calculating temperature trends which you believe implicitly.
For a different discipline its OK for you accept anomalies, but not for economics? OK. Got it!
CoRev: One type of statistical mean is an arithmetic average (there are also geometric averages, and harmonic averages, etc. etc.). I have no clue what you are trying to get at.
Now, one can always calculate an arithmetic average for a sample. We do that typically because we want an estimator of the population mean. That is why there is a standard error of estimate for an arithmetic average. That arithmetic calculation has a meaning if there is a true population mean. If there isn’t then the calculation is not useful for thinking about the population mean. If a series is an I(1) process, then the variable does not have a population mean.
I don’t know how much more clearly I can explain this to someone who clearly has not absorbed any information from basic statistics.
Menzie, how many times are you going to say: ” I have no clue what you are trying to get at”? You have shown you also do not know about anomalies used with temperature or averages used with calculating anomalies or even arithmetic averages for nonfarm payroll employment .
I guess you are just looking at starting another soybean saga.
Menzie,
Oh dear, there you go again, trying to explain basic time series statistics to CoRev. But no matter how right you are, it does not matter. He has you! I mean, he imitated Moses Herzog by emboldening his quotation from you where you admitted you did not understand what he was doing! Victory for CoRev, so emphasizing your inability to understand him!
And now he has brought out not just one new goal post, but potentially a whole swamp full of them by mentioning the “soybean saga,” which he clearly thinks was another great victory for him. Gotha again!
Now, speaking of really important goal posts, all we need is for him to tell us just exactly which award that he put on his wall from his profoundly important service in the Apollo program he was most proud of. That will really clarify exactly what it is that he has been trying to do here, as the leading economist and climatologist on this blog for sure.
Wow! Barkley not only changed the goal post he changed the game, location, and time frame: “Now, speaking of really important goal posts, all we need is for him to tell us just exactly which award that he put on his wall from his profoundly important service in the Apollo program he was most proud of. That will really clarify exactly what it is that he has been trying to do here, as the leading economist and climatologist on this blog for sure.”
As he says – gag!
Hey, CoRev, if you can drag in the “soybean saga,” a major embarrassment for you, and also make up wild claims that 2slug wants to spend trillions of dollars a year he never said anything about, well I can mention your embarrassing rantings about your awards. I mean don’t you have so many medals if you wear them all you will fall over because you have more than some leftover WW II Soviet general?
So, I suggest that just to show you are really on top of all this, I suggest you use the word “gibberish” a few more times about something somebody else said, embolden your most important statements such as your obviously killer comment about how poor Menzie is just so hopeless because he has said he does not understand what you say, and, oh yes, call pgl “Barkling Bierka” a few more times, that one is really impressive.
Reader CoRev has provided his check-file on my trend analysis
Oh my. That’s embarrassing.