The Research Program on Forecasting supports research, teaching, and dissertation supervision in forecasting as part of the Department of Economics and the Center for Economic Research at The George Washington University. The current research interests of program members include a wide range of studies on the methodology of forecasting and forecast evaluation as well as preparation of macroeconomic and microeconomic forecasts. In addition, program members have supervised dissertations that focus on the theory and application of forecasting.

Announcing the 17th Federal Forecasters Conference, 
to be held Thursday, September 24, 2009. 
The theme will be "Forecasting and Risk." 
Click here for a call for papers.

Brown Bag Seminar Series on Forecasting
(joint with the Federal Forecasters Consortium)

12:30 pm - 2 pm
John W. Kendrick Seminar Room
Room 321
2115 G Street
Washington, DC 20052

Upcoming Presentations
Note: We are currently on summer break.  Look for new seminars scheduled for this fall!
Past Presentations
Tuesday, May 12, 2009: Jaime Marquez
Division of International Finance, Federal Reserve Board
Measuring U.S. International Relative Prices: A WARP View of the World

Tuesday, April 14, 2009: Kajal Lahiri
Measuring Forecast Uncertainty by Disagreement: The Missing Link

Special Forecasting Seminar: Friday, March 13, 2009, 2:30 to 4:00 pm
Ralph Snyder, Department of Econometrics and Business Statistics
Monash University, Australia
Rethinking our Approach to Time Series Analysis
Click Here for the Handout

Abstract:  Those of us trained in time series analysis in the latter half of the last century were weaned on the Box-Jenkins approach, so much so that it became the dominant way of thinking in areas such as time series econometrics, despite other developments such as exponential smoothing and the state space approach. It will be argued in this presentation that we need to rethink how we do time series analysis to avoid subtle but important pitfalls associated with the traditional Box-Jenkins approach. I will suggest that a single source of error state space approach avoids these pitfalls and should therefore replace  the Box-Jenkins approach as the dominant framework for time series analysis.

Tuesday, March 10, 2009: Fred Joutz
“Modeling and Forecasting Electricity Loads: Capturing Hourly, Daily, and Seasonal Effects”

Tuesday, February 10, 2009:
Dean Croushore, University of Richmond and Interim Director, Real-Time Data Research Center, Federal Reserve Bank of Philadelphia
“Evaluating Inflation Forecasts in Real Time.”

Tuesday, January 13, 2009:
Jeffrey F. Werling, Inforum
"Forecasting" the U.S. Economy Over the Very Long Term

Thursday, December 11, 2008: 
Chris Varvares, President, Macroeconomic Advisers
Macroeconomic Outlook

Special Joint Seminar with the GWU Macro-International Seminar Series:
Wednesday, November 19, 2008 at 2:30 pm
James Morley, Washington University in St Louis
"The Asymmetric Business Cycle" (with Jeremy Piger)

Tuesday, November 18, 2008:
Prakash Loungani, IMF and RPF
"Housing Prices: Global Correction, Local Consequences" (joint work with Charles Collyns, Marcello Estevao and Deniz Igan).

Thursday, September 18, 2008:  Turgut Kisinbay, IMF
"The Use of Encompassing Tests for Forecast Combinations"

Thursday, May 22, 2008:  Kathryn Byun, BLS
"The Housing Bubble and the Resulting Impact on Employment"

From the late 1990s through 2006, investment in residential construction grew at an unprecedented rate.  This demand resulted in jobs not only in the construction industry but throughout the economy.  Using the input output system developed by the Employment Projections program within the Bureau of Labor Statistics, we can estimate the employment by both industry and occupation that was generated to meet this run-up in demand.  Historical data from 1996-2006 and projected data for 2016 will be examined for residential and nonresidential construction and compared to overall demand. 
Thursday, April 24, 2008:  FFC Conference!

Thursday, March 13, 2008:  Ed Gamber, Lafayette College
"An Absorption Approach to Modeling the U.S. Current Account,"
co-authored with Juann Hung, Congressional Budget Office.

This paper derives and estimates a current account model from the absorption approach, which views the current account balance as the difference between domestic saving and investment.  This approach provides a framework which allows determinants of savings, investment, and capital flows to be included in a current account model, an advantage not offered by the elasticity approach, which views the current account balance as the sum of net exports and net investment income.  We estimate and compare vector error correction models of the absorption and elasticity approach.  The restrictions on the model imposed by the elasticity approach are rejected using quarterly data from 1973 through 2007.  In addition, we find that the out-sample forecasts of a current account equation based on the absorption approach perform better than those of an equation based on the elasticity approach.

Thursday, February 21, 2008: Prakash Loungani, IMF
"Coupled Economies, Decoupled Forecasters?" 

Thursday, January 17, 2008: Stephen MacDonald, Senior Economist, Economic Research Service, USDA
"Integrating Judgemental and Quantitative Forecasts"
Traditional methods of forecasting international commodity trade are no longer feasible for the Department of Agriculture.  Declining human resources, and the increased availability of timely trade data are forcing a transition to more quantitative approaches from the traditional system of reports from overseas embassies.  This study examines how empirical confidence intervals can help bridge the gap between USDA’s judgemental, Delphic approach and the efficient application of growing quantitative resources.

Thursday, November 8, 2007:  Julie Smith, Lafayette College
"Has the Fed's Forecasting Advantage Eroded?" co-authored with Ed Gamber.

Thursday, October 18, 2007:  Tara Sinclair
"Jointly Evaluating the Federal Reserve’s Quantitative Forecasts of Real GDP Growth and Inflation," co-authored with Herman Stekler and Elizabeth Reid.

Tuesday August 7, 2007:  Antony Davies
"A Framework for Decomposing Shocks and Measuring Volatilities Derived from Multi-Dimensional Panel Data of Survey Forecasts"
12:00 - 1:30 pm
John W. Kendrick Seminar Room
Room 321
2115 G Street
Washington, DC 20052

Tuesday, June 19, 2007:  Stefan Osborne
"Cellulosic Ethanol: Effects on the Future U.S. Economy of Successful Commercialization"

April  17, 2007:  Herman Stekler
(joint with the GWU Microeconomics Seminar)
"A Survey of Results from Sports Forecasts" new version 8/13/07!

March 20, 2007:  Kathleen Sorensen
“Forecasting Interments and Gravesites in National Cemeteries, 
the New Model”

February 20, 2007:  Herman Stekler
"Evaluating BLS Labor Force, Employment, 
and Occupation Projections for 2000" 
by H.O. Stekler and Rupin Thomas.

January 9, 2007:  Tara M. Sinclair
"Directional Forecasts of GDP and Inflation: A Joint Evaluation 
With an Application to Federal Reserve Predictions"
by Tara M. Sinclair, H.O. Stekler, and Lindsay Kitzinger.
paper link presentation slides link


Current Activities

The faculty are all active scholars. They have published in some of the top general interest journals as well as leading journals in forecasting, including the American Economic Review, Journal of the American Statistical Association, Journal of Business and Economic Statistics, Economic Inquiry, Journal of Money Credit and Banking, International Journal of Forecasting, Applied Economics, Journal of Applied Econometrics, andEnergy Economics. Their work also appears in leading journals in related fields and disciplines, (such as macroeconomics and demography), and they regularly present their work at national and international forecasting conferences.

Ord and Stekler are members of the editorial board of the International Journal of Forecasting. Stekler and Ord are both former Directors of the International Institute of Forecasters and are Fellows of the Institute.  Ord is also a Fellow of the American Statistical Association.

The Survey of Professional Forecasters conducted by the Federal Reserve Bank of Philadelphia has included the Benchmark Forecasts produced by Joutz since 1990. He has been producing the forecasts of about 25 macroeconomic series since 1988. The survey is done every quarter and includes about 35 forecasters from business, finance, and aceademe. Previously, the survey was conducted under the auspices of the ASA and National Bureau of Economic Research.  Tara Sinclair also contributes to the survey on behalf of the Research Program on Forecasting.

Previously funded research projects include a three year project on short-run and long-run responses of electricity consumption to climate change and variability for the U.S. Environmental Protection Agency. The project  was performed in collaboration with scientists at the Johns Hopkins Schools of Engineering and Public Health Other funds have come from the National Park Service.

The RPF also has a partnership with Federal Forecasters Consortium, an organization of federal employees from all government agencies involved in forecasting. The two groups jointly organize a monthly Brown Bag Seminar Series on Forecasting.


Research

Specific projects underway or recently completed by program members include:

Recent Expert Testimony

Recent Audio and Video
  • November 24, 2008: Tara Sinclair was interviewed by Fox 5 morning news about the economic outlook.
  • November 13, 2008: Tara Sinclair was interviewed by Fox 5 morning news about the modifications to the bailout plan.
  • October 22, 2008:  Tara Sinclair was interviewed by Medill Reports about the different presidential candidates' tax policies.
Other Recent Press:
  • April 6, 2009: A Bloomberg article discussed a working paper by Tara Sinclair. 
  • August 17, 2008:  The work presented by Prakash Loungani in our February brownbag was mentioned in the New York Times

  • Magazine in an article about Nouriel Roubini:
    "Recessions are signal events in any modern economy. And yet remarkably, the profession of economics is quite bad at predicting them. A recent study looked at "consensus forecasts" (the predictions of large groups of economists) that were made in advance of 60 different national recessions that hit around the world in the '90s: in 97 percent of the cases, the study found, the economists failed to predict the coming contraction a year in advance. On those rare occasions when economists did successfully predict recessions, they significantly underestimated the severity of the downturns. Worse, many of the economists failed to anticipate recessions that occurred as soon as two months later."
    "He sounded like a madman in 2006," recalls the I.M.F. economist Prakash Loungani, who invited [Nouriel] Roubini [to the IMF] on both occasions.
    "He was a prophet when he returned in 2007."
  • Prakash Loungani blogs on the excessive smoothness of forecasts for RGE Monitor:

  • "Think about it this way: what use is a weather forecaster who never calls for rain until both you and he can look out of the window and see that it is indeed raining? A weather forecaster is doing his job if he does call, every now and then, for rain ahead of time, even at the risk of being wrong and annoying his clients for making them carry an umbrella around or rescheduling outdoor tennis matches when they didn’t have to. A weather forecast that bounces around can be annoying but it may also be a sign of the forecaster taking his job seriously."
  • WSJ.com blogs about Ed Gamber and Julie Smith's Working Paper (Working Paper No. 2007-002 below).
Recent Publications:
  • Are the Fed’s Inflation Forecasts Still Superior to the Private Sector’s? by Edward N. Gamber and Julie K. Smith, RPF Working Paper No. 2007-002, is now forthcoming in the Journal of Macroeconomics.
  • "Economic Forecasts: Too Smooth By Far?" by Prakash Loungani and Jair Rodriguez in now published in World Economics.

  • Summary: Will the U.S. economy go into a recession? Will other countries “decouple” from the United States? To those turning to private sector economic forecasters for guidance on these questions, this article sounds a note of caution: the past performance of forecasters on these two questions leaves much to be desired. Few recessions have been forecast ahead of their arrival.Nor do forecasters take into account fully the dependence of economies on one another; we present evidence that the dependence on the U.S. economy of the other members of the G7 has been under-appreciated by forecasters in the past. In short, economies are coupled, but one country’s forecasters often seem to be decoupled from those in other countries. 
  • "Directional Forecasts of GDP and Inflation: A Joint Evaluation With an Application to Federal Reserve Predictions" by Tara M. Sinclair, H.O. Stekler, and Lindsay Kitzinger is now forthcoming in Applied Economics.
  • "An Evaluation of the Forecasts of the Federal Reserve:  A Pooled Approach" by Michael P. Clements, Fred Joutz, and Herman O. Stekler is now published in the Journal of Applied Econometrics 22(1), 121-136. 
Recent Presentations: Forecast evaluations: 
  • Federal Reserve Board forecasts of macroeconomic conditions 
  • Expert, computer model, and betting line forecasts of NFL games 
  • Interest rate spreads as predictors of business cycles 
  • U.S. Census Bureau projections of state populations 
  • The role of asymmetric, non-linear models in forecasting 
Preparation of forecasts of:
  • U.S. macroeconomic conditions 
  • Electricity consumption and prices 
  • Motor gasoline demand 
  • Municipal solid waste generation 
  • Effects of technical change on productivity in public enterprises 
  • Patent applications 
  • Leading series of the transportation sector 
  • National defense readiness measures

Members

Frederick L. Joutz, Professor of Economics, Program Director, and member of the Federal Forecasters Consortium Governing Board.
Tara M. Sinclair, Assistant Professor of Economics and International Affairs, Program Co-Director, and member of the Federal Forecasters Consortium Governing Board.
Bryan L. Boulier, Professor of Economics
Michael D. Bradley, Professor of Economics
Antony Davies, Associate Professor of Economics, Duquesne University
Edward N. Gamber,  Professor of Economics, Lafayette College, and former Macroeconomic Forecaster of the Congressional Budget Office 
Prakash Loungani, Advisor in the Research Department of the IMF (International Monetary Fund)
Michael W. McCracken, Research Officer, Federal Reserve Bank of St. Louis
Robert Phillips, Professor of Economics, Chair
J. Keith Ord, Professor, McDonough School of Business, Georgetown
University, Fellow of International Institute of Forecasting, Fellow of
American Statistical Association 
Refik Soyer, Professor of Management Science 
Julie K. Smith, Assistant Professor of Economics, Lafayette College
H.O. Stekler, Research Professor of Economics and Fellow of International Institute of Forecasting
Peter Tinsley, Professor of Economics, Birkbeck College, University of London
Robert Trost, Professor of Economics
Anthony M. Yezer, Professor of Economics


Contact Information

Fred Joutz 
Department of Economics 
Monroe/Hall of Goverment Suite 313
2115 G Street NW
Washington, DC 20052 
(202) 994-4899 
fax:  (202) 994-6147 
e-mail: bmark@gwu.edu


Forecasting Links

Working Papers 

2009
No. 2009-001: Can the Fed Predict the State of the Economy?  Tara M. Sinclair, Fred Joutz, and Herman O. Stekler.
Note: This is a substantially revised version of RPF Working Paper No. 2008-010.

2008

No. 2008-011: Bootstrap Tests of Stationarity James Morley and Tara M. Sinclair.

No. 2008-010: Are 'unbiased' forecasts really unbiased? Another look at the Fed forecasts  Tara M. Sinclair, Fred Joutz, and Herman O. Stekler.

No. 2008-009: What Do We Know About G-7 Macro Forecasts? Herman O. Stekler.

No. 2008-008: Exhaustive Regression: An Exploration of Regression-Based Data Mining Techniques Using Super Computation Antony Davies.

No. 2008-007: Evaluating Census Forecasts Herman O. Stekler.

No. 2008-006: Measuring Consensus in Binary Forecasts: NFL Game Predictions ChiUng Song, Bryan L. Boulier, and Herman O. Stekler.

No. 2008-005: Evaluating Current Year Forecasts Made During the Year:
A Japanese Example H.O. Stekler and Kazuta Sakamoto.

No. 2008-004: Monitoring Processes with Changing Variances J. Keith Ord

No. 2008-003: Exponential smoothing and non-negative data Muhammad Akram, Rob J Hyndman, and J. Keith Ord.

Updated 8/8/08! No. 2008-002: Multivariate Forecast Errors and the Taylor Rule  Edward N. Gamber, Tara M. Sinclair, H.O. Stekler and Elizabeth Reid.

Updated 3/20/2009! No. 2008-001:   Forecast Errors Before and After the Great Moderation Edward N. Gamber, Julie K. Smith, and Matthew Weiss.

2007

Updated! 8/6/2008: No. 2007-002: Are the Fed’s Inflation Forecasts 
Still Superior to the Private Sector’s? Edward N. Gamber and Julie K. Smith.   Note:  A revised version of this paper is now forthcoming in the Journal of Macroeconomics!

No. 2007-001: Sports Forecasting  Herman O. Stekler.

2006

No. 2006-001: An Evaluation of the Forecasts of the Federal Reserve:  A Pooled Approach. Michael P. Clements, Fred Joutz, and Herman O. Stekler.  Note:  A revised version of this paper is now published in the Journal of Applied Econometrics 22(1), 121-136.

No. 2006-002 updated 11/14/06: Directional Forecasts of GDP and Inflation: A Joint Evaluation With an Application to Federal Reserve Predictions Tara M. Sinclair, H.O. Stekler, and Lindsay Kitzinger.  Note:  A revised version of this paper is now forthcoming in Applied Economics.
 


Research Program on Forecasting/ bmark@gwu.edu

Revised: May 6, 2009