I am constantly amazed that people write stuff that is easily falsifiable, in an era of easily accessible databases (I am tempted to go into an old fogey rant about “in my day I had to go to the library and hand copy down numbers from the hard copy volumes of International Financial Statistics”…but I will resist). Or ask me for the “raw” data when it’s freely available.
Just to remind the frequent commenters on this blog, there exist freely available data here:
- St. Louis Fed economic database Thousands of time series on economic activity, in an easily downloadable form.
- IMF International Financial Statistics
- IMF World Economic Outlook databases.
- World Bank World Development Indicators.
- DBnomics (a “European FRED”)
- YCharts Macro and equity market data series.
- ino.com Futures data.
- Federal Reserve Board data Monetary, financial and output data collected by the Nation’s central bank.
- Bureau of Economic Analysis, Dept. of Commerce Data on GDP and components (the national income and product accounts) as well as other macroeconomic data.
- Bureau of the Census, Dept. of Commerce Data on the characteristics of the US population as well as of US firms.
- Bureau of Labor Statistics, Dept. of Labor Data on wages, prices, productivity, and employment and unemployment rates.
- Energy Information Agency, Dept. of Energy Data on energy (electricity, gas, petroleum) production, consumption and prices.
- Economic Report of the President, various years. The back portion of this annual publication contains about 70 tables of government economic data.
- Economic Indicators CEA and JEC Compilation of economic data in tabular form.
- Economic Time Series page A large collection of economic time series.
- NBER Data Specialized economic databases created by economists associated with the National Bureau of Economic Research.
- Penn World Tables.
- Netherlands Bureau for Economic Policy Analysis World Trade Monitor
I have readers on Econbrowser pestering me for “raw” data used. Usually I cite FRED or BLS via FRED, or from the above data sources. In certain cases, I have written papers using specialized data sources. Below are links to those data sources.
- “A Faith-based Initiative: Do We Really Know that a Flexible Exchange Rate Regime Facilitates Current Account Adjustment,” Review of Economics and Statistics (March 2013) (with Shang-Jin Wei) [PDF]. Data: Dataverse
- “A New Measure of Financial Openness,” Journal of Comparative Policy Analysis 10(3) (September 2008): 307-320 (with Hiro Ito) [PDF]. Chinn-Ito index used in World Bank, World Development Report 2009. Data: Chinn-Ito capital account index webpage
- Financial Spillovers and Macroprudential Policies,” mimeo (with Joshua Aizenman and Hiro Ito). Under revision. [PDF] ; “Balance Sheet Effects on Monetary and Financial Spillovers: The East Asian Crisis Plus 20,” paper presented at JIMF-University of Tokyo Conference “The Pacific Rim and the Global Economy: Future Financial and Macro Challenges” July 25 and 26, 2016.(with Joshua Aizenman and Hiro Ito). [PDF] Data: Aizenman-Chinn-Ito Trilemma indices webpage
- “The Predictive Power of the Yield Curve across Countries and Time,” International Finance (March 2015) [PDF] Cited in “Free Exchange: Bond yields reliably predict recessions. Why?”Economist (July 26, 2018) Data (Stata file): Chinn-Kucko data
- “Post-recession US employment through the lens of a non-linear Okun’s law,” Journal of Macroeconomics 42 (December 2014): 118–129, with Laurent Ferrara and Valérie Mignon. [PDF] Data: data [XLS] RATS program
- “The Predictive Content of Commodity Futures,” Journal of Futures Markets, July 2014 (with Olivier Coibion). [PDF] Data: Data [xlsx]
- “Monetary Policy and Long-Horizon Uncovered Interest Rate Parity,” IMF Staff Papers 51(3) (2004) (with Guy Meredith) [PDF] Data: [Excel file]; Notes [PDF].
And from teaching:
You are just a young whippersnapper. I am old enough to remember creating data series from scratch gathered from index cards and scattered reports in file cabinet drawers of government agencies, writing it all own by hand. Hand copying out of library copies of already gathered series? Hah! Wussie child’s play.
I was recently remembering carrying around a kit of drawing supplies and a drawing board for “engineering drawing lab….. what are all those draftsman doing (since the ’80’s) today?
As always a futile effort.
Who you gonna believe, my biased politics or the lying data???
“I am constantly amazed that people write stuff that is easily falsifiable, in an era of easily accessible databases … I have readers on Econbrowser pestering me for “raw” data used. Usually I cite FRED or BLS via FRED, or from the above data sources.”
Ah but you don’t get these readers motives. With CoRev – unless it is preapproved intellectual garbage from the Trump White House then it is fake data. And your data has not gotten you invited on Fox & Friends the way Princeton Stevie boy has. After all – he just makes up his data to please the right wingers who ran those shows.
You are right that this isn’t and shouldn’t be “required” of you. However it’s a kind and gracious gesture, it’s nice to have the data encapsulated as you do here where people can “bookmark” it etc on their computer. It’s a classy thing to do and the type of thing a great professor does, There is an informal term in college football called “a player’s coach”. Barry Switzer was always referred to as “a player’s coach”. Buddy Ryan and Rex Ryan were also called “a player’s coach”. When I was a pretend teacher I tried to exemplify these qualities to my students (sometimes failing, but somewhere in my brain with that end aim).
I went to college in the mid ’90s and the net was JUST starting to come in then. At that time Yahoo and Netscape were commonly used, and Netscape was light years ahead of the microsoft browser at that time. It was where you might use the internet for maybe 1/8 to 1/4 of your studies and the rest you would do the old fashioned way, hunting down volumes or using a type of card catalogue or books system to hunt obscure periodicals. If you loved learning, flipping through those things or using a ladder to reach the top part of the shelves could be a kind of “labor of love” when you’re young and have energy. Damned whipper-snappers!!!!!
*All background research on the commodity of orange juice in the above video done by CoRev.
There was a French one too that you highlighted wasn’t there, Menzie?? I will check my RSS reader here for the name “DB nomics”?? I guess good for some of the European numbers
I just noticed they have an “R” client and a Python client. That is pretty BADASS!!!! Gretl and Julia, I think Julia is very new and is supposed to be some kind of “AI” type software from Massachusetts Institute of Technology. That is super super cool. I feel a tinge stupid because all this time I thought “DB” implied Deutsche Bank.
I thank you for posting data in the past and for posting a few EViews models in the past. The postings have been very helpful to learn more about time series modeling and forecasting. A couple years ago, you posted some cointegration models and related Johansen tests. These were very helpful. If I am one of the culprits, my apology is offered.
AS: No, you have been the ideal blog commenter. I applaud your desire to learn.
Time series econometrics is an unusual hobby. What motivated you to get into this?
@ Rick Stryker
The desire for knowledge and self-education. Most likely an emotion you’ve never felt before.
Think about when you’re hungry for dessert—that’s probably the closest you’ll ever come to grasping the internal feeling.
One of my first jobs after graduate school and after audit experience with the Big Eight was as budget director for a public hospital company. As I recall we had around 15 or 20 hospitals in multiple states. A key activity and revenue driver for hospital budgets at that time was patient days. I wanted a method of forecasting patient days to distribute to each hospital that had some chance of being believable and that would not take me untold hours of toil to forecast. I reviewed the programs available on the company’s time-share computer terminal (no PCs at the time) and found a Holt-Winters program. Since hospital volume is highly seasonal, the HW program worked well. Moving on to management positions I was no longer involved with forecasting, so further study had to wait until I retired. The software now available for use on PCs and some of the books now available allow for self study. One book that is accessible for those without advanced economics or math degrees is, “Forecasting for Economics and Business” by Gloria Gonzalez-Rivera (who I think was a student of Professor Hamilton). Professor Dave Giles has also provided examples that have been very useful on his blog. Each week economic indicators are announced, so it is fun to try to match wits with the experts. FRED offers a monthly forecasting game that is very challenging to score in the top 10%. Forecasting also, helps to keep informed about what is happening in the economy. I know a thimble-full compared to the expert readers, but still have fun trying to understand what seems to be an endlessly complex subject.
That’s crazy man. I been taking an online course led by CoRev on soybeans. Here’s a video of “Unit 1” of the course, so far it’s cost me $19.95 (an offer only available for a limited time) and my real life results have been, uuuuuhh, well I would say “mixed” results.
I needed a change in my life and it was either CoRev’s online course on soybeans or Joel Osteen, and I was already fatigued with Osteen’s whispery sound by the 3rd day.
This guy AS has a very high midi-chlorians count if you ask me:
Kinda reminds me of this chick.
Just one note about the data. If you are going to use the Energy Information Administration stuff, you want to be very careful about which series you use, and citing it. All these little meticulous definitions matter. Let me give you one current example. It is from
which has the data for April 2019. You look at the electricity prices by state in
Table 5.6.A. Average Price of Electricity to Ultimate Customers by End-Use Sector,by State, April 2019 and 2018 (Cents per Kilowatthour)
You can say electricity prices in Maryland, between April 2018 and April 2019, went up from 13.37 cents per Kwh to 14.25 cents per Kwh, a 6.58% increase, if you want to worry about inflation, or you can say they went up from 11.54 cents per Kwh to 11.60 cents, a 0.52% increase, if you are confident inflation is not a problem. It takes careful digging to realize that the first is referring to residential prices exclusively, while the second has all prices, weighted by quantities used. The second includes industrial prices, which fell, from 8.35 cents per Kwh in April 2018 to 7.66 cents per Kwh in April 2019. These sorts of irritating details can be overlooked, shall we say, by people who are confident they are right and don’t need to worry.
Julian Silk: Yes, that is a good point. The bottom line: always read the footnotes, when it comes to data.
Another note about your data – do you think the “Big Mac” index is a good way to compare purchasing power parity exchange rates? I believe the index is created using prices after all relevant local taxes. But the index makes no allowance for value added taxes, which usually are adjusted at the border using the destination principle. Value added taxes are not uniform among countries. Although most U.S. states have sales taxes that are effectively rebated at the border, their average level is far below the value added taxes in most of Europe. For example, an average VAT of 25% and sales taxes of 6% among the U.S. states would cause the index to overstate substantially the purchasing power parity value of the dollar vis-a-vis the euro.
don: As I recall the literature on burgernomics, most of the cost was in labor. Hence, I don’t think the bias is as great as you assert. Probably dwarfed by the exchange rate volatility.