Each month I have been highlighting various estimates of U.S. recession probabilities. The latest update was that of April 6, 2016, titled "Recession Probability Models - April 2016."
While I don't agree with the methodologies employed or the probabilities of impending economic weakness as depicted by these and other estimates, I do believe that the results of these models and estimates should be monitored.
Another probability of recession is provided by James Hamilton, and it is titled "GDP-Based Recession Indicator Index." A description of this index, as seen in FRED:
This index measures the probability that the U.S. economy was in a recession during the indicated quarter. It is based on a mathematical description of the way that recessions differ from expansions. The index corresponds to the probability (measured in percent) that the underlying true economic regime is one of recession based on the available data. Whereas the NBER business cycle dates are based on a subjective assessment of a variety of indicators that may not be released until several years after the event , this index is entirely mechanical, is based solely on currently available GDP data and is reported every quarter. Due to the possibility of data revisions and the challenges in accurately identifying the business cycle phase, the index is calculated for the quarter just preceding the most recently available GDP numbers. Once the index is calculated for that quarter, it is never subsequently revised. The value at every date was inferred using only data that were available one quarter after that date and as those data were reported at the time.If the value of the index rises above 67% that is a historically reliable indicator that the economy has entered a recession. Once this threshold has been passed, if it falls below 33% that is a reliable indicator that the recession is over.
Additional reference sources for this index and its construction can be seen in the Econbrowser post of February 14, 2016 titled "Recession probabilities" as well as on the "The Econbrowser Recession Indicator Index" page.
Below is a chart depicting the most recent value of 15.70%, for the fourth quarter of 2015, last updated on April 28, 2016 (the date of the Gross Domestic Product Q1 2016 Advance Estimate (pdf):
source: Hamilton, James, GDP-Based Recession Indicator Index [JHGDPBRINDX], retrieved from FRED, Federal Reserve Bank of St. Louis on April 2, 2016:
_____
The Special Note summarizes my overall thoughts about our economic situation
SPX at 2081.43 as this post is written
No comments:
Post a Comment