Entering text into the input field will update the search result below

Charts Of The Day: 6 New Record Highs - U.S. Breweries, World Stock Market Cap, World Trade, World Output, U.S. Gas, U.S. Oil

Mark J. Perry profile picture
Mark J. Perry

This article was written by

Mark J. Perry profile picture
Dr. Mark J. Perry is a full professor of economics at the Flint campus of The University of Michigan, where he has taught undergraduate and graduate courses in economics and finance since 1996. Starting in the fall of 2009, Perry has also held a joint appointment as a scholar at The American Enterprise Institute. Perry holds two graduate degrees in economics (M.A. and Ph.D.) from George Mason University and in addition, and has an MBA degree in finance from The University of Minnesota. In addition to an active scholarly research agenda, Perry enjoys writing op-eds for a general audience on current economic issues and his opinion pieces have appeared in most major newspapers around the country, including USA Today, Wall Street Journal, Washington Post, Investor’s Business Daily, The Hill, Washington Examiner, Dallas Morning News, Sacramento Bee, Saint Paul Pioneer Press, Miami Herald, Pittsburgh Tribune-Review, Detroit News, Detroit Free Press and many others. Mark Perry has been best known in recent years as the creator and editor of one of the nation’s most popular economics blogs, Carpe Diem. Professor Perry has written on a daily basis since the fall of 2006 to share his thoughts, opinions and expertise on economic issues, with a strong emphasis on displaying economic data in a visually appealing way using graphs, charts and tables.

Recommended For You

Comments (1)

Oil weeklies are 95% a model and 5% data. They take the STEO FORECAST and just construct weekly estimates by linearly interpolating the STEO monthly FORECAST. The only new data in there is the AK pipe flow (~500 M bpd).

Weeklies are also never revised for bad old values (so it is not a curated time series). Instead each week is just a best guess, really a forecast of current production. There are big jumps on a monthly basis when new (2 month old) data from the 914 is obtained and changes the STE forecasting.

For all these reasons, it is a bad idea to show weekly values as a time series or even to cite recent weekly numbers as an EIA report (as if it were survey data). The 914 report and resultant time series is what you should plot instead. I have told this to Perry before but he just repeats the error (to the extent of doing advocacy, not analysis). He might have the same politics as Milton F., but not anything like some analytical rigor.
Disagree with this article? Submit your own. To report a factual error in this article, . Your feedback matters to us!
To ensure this doesn’t happen in the future, please enable Javascript and cookies in your browser.
Is this happening to you frequently? Please report it on our feedback forum.
If you have an ad-blocker enabled you may be blocked from proceeding. Please disable your ad-blocker and refresh.