In Memory of Herman Stekler

September 5, 2018
Herman Stekler in 2015 with the issue of the International Journal of Forecasting that included a special section in his honor.

Herman Stekler in 2015 with the issue of the International Journal of Forecasting that included a special section in his honor.

Herman Stekler, a giant in the field of forecasting and a dedicated mentor and teacher, died at home on Tuesday, September 4, just two months shy of his 86th birthday. Herman served as an associate editor of the International Journal of Forecasting (IJF) from 1987 through the January-March 2018 Issue. Herman was also book review editor for the IJF, a director of the International Institute of Forecasting (IIF) and an Honorary Fellow of the IIF. With well over 100 papers, Herman is ranked in the top 5 percent of RePEc authors.

The majority of Herman’s research was focused on the production and evaluation of macroeconomic forecasts. His JPE article with Sidney Alexander 1959 outlined the issues macroeconomic forecasters and evaluators still face today. Herman also wrote a number of papers on sports forecasting, including “Are sports seedings good predictors?: an evaluation,” in the International Journal of Forecasting in 1999 (joint with his colleague at the George Washington University, Bryan Boulier). Herman evaluated the Federal Reserve’s forecasts from a number of different dimensions and was most concerned that the Fed consistently missed recessions. One of Herman’s most cited works according to RePEc is “An evaluation of the predictions of the Federal Reserve” (joint with another of his colleagues at the George Washington University, Fred Joutz) in the International Journal of Forecasting in 2000. Herman’s research provided new tools in forecast evaluation with a focus on turning points, qualitative forecasts, and joint forecasts. He also advocated for more research focused on predicting recessions by improving leading indicators and diffusion indexes. 

According to his IIF member profile, Herman did not plan on becoming a forecaster when he started his PhD studies at MIT in 1955, but Sidney Alexander joined the MIT faculty in 1956 and convinced Herman to work with him on indicator forecasting. After graduate school Herman took a faculty position as an assistant professor of business administration at Berkeley and explored changing his research focus to industrial organization. Fortunately for forecasting Herman was offered a consulting project to “evaluate the forecasting record of all existing econometric models.” The project brought him into research on forecast evaluation, the field he led and added new knowledge to through his last months.

Stekler left Berkeley to be an economist at the Federal Reserve Board from 1966-68, where he was involved in the early Greenbook forecasts that he later evaluated from many different dimensions in his research. He was recruited to a full professor position in the department of economics at the State University of New York Stony Brook from 1968 through 1977. When his wife, Lois, received an offer to return to the International Finance Division at the Federal Reserve Board, he supported her career move and came to Washington. There he was appointed to serve on President Carter’s Council on Wage and Price Stability (COWPS), before moving to the Industrial College of the Armed Forces (ICAF) where he worked until he retired in 1994.

Retirement for Herman did not mean leaving research or teaching, however. Instead he joined the George Washington University as a research professor and a member of the GW Research Program on Forecasting where he continued to produce cutting edge research and teach undergraduate and graduate classes. He was in his GW office nearly every day from 1994 through 2017 and published over 20 papers with GW students.

Herman also continued to serve the policy community such as forecast evaluations for the Employment Projections program at Bureau of Labor Statistics. According to an economist there at the time, his “contribution led to comparison against a benchmark projection (a naïve forecast) that added rigor to the comparisons.”

Herman will be remembered as one of the founding fathers of forecasting as well as the “grandfather” of forecast evaluation. The IJF published an interview of Herman Stekler by Fred Joutz in 2010 and a special section of the journal in honor of Herman Stekler in 2015 (see Rob Hyndman’s blog post for an introduction to the section).

The GW Research Program on Forecasting plans to rename itself the H.O. Stekler Research Program on Forecasting in Herman’s memory. A fund has been set up to support this renaming initiative. Following Herman’s wishes, the funds raised will support student engagement in forecasting research.

Written by Fred Joutz, Prakash Loungani and Tara Sinclair.