A methodologist's econometric model selection

  • Brigitte H. Bechtold


The use of shortcuts to predict time series is discouraged by the findings of this paper. First, if time series data are no longer considered to be the result of random sampling, statistical analyses based on properties of sampling distributions become questionable. Using the shoal of fish analogy, we should be wary of considering our catch of fish caught in one dip of the net as essentially identical to another catch obtained from another dip. Second, transformation of time series data to achieve stationarity prevents the user of VAR-related methods from relying on information that is potentially symptomatic of misspecification, especially in the form of missing variables. Since the incidence of missing variables is enhanced the more “compact” is the model used, the implications of the stationarity assumption are vast. Third, uninformed comparisons of the short-term forecasting performance of VAR and large-scale commercial forecasting models leads to a more positive evaluation of VAR than is otherwise warranted. Overall, using our shoal of fish metaphor one final time, we should focus more attention on the reasons why the shoal is in its present location, the reasons why the fish are the size they appear to have, the reasons why the size of the shoal increases and the reasons why the shoal moves from one location to another. If there are sharks around, we should want to know, rather than ignoring their presence.


Time Series Economic Growth Random Sampling Model Selection Series Data 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Albert Ando and Brigitte H. Bechtold,On the Tight Monetary Policy of 1974. University of Pennsylvania, Department of Economics, 1975. (names listed as Ando and Sellekaerts).Google Scholar
  2. Brigitte H. Bechtold, "Effect of the Minimum Wage on Inflation and Other Key Macroeconomic Variables,"Eastern Economic Journal, 3(3), July 1982, pp. 177–190. (name listed as B. Sellekaerts).Google Scholar
  3. G.E.P. Box and G. M. Jenkins,Time Series Analysis: Forecasting and Control. San Francisco: Holden Day, 1970.Google Scholar
  4. R. Bentzel and H. Wold, "On Statistical Demand Analysis from the Viewpoint of Simultaneous Equations,"Skand. Actuarietidskr., 29 (1946), pp. 95–114.Google Scholar
  5. Adrian C. Darnell and J. Lynne Evans,The Limits of Econometrics. Brookfield, Vt: Gower Publishing Company, 1990.Google Scholar
  6. Robert F. Engle and C. W. J. Granger, eds,Long-Run Economic Relationships; Readings in Cointegration; New York: Oxford University Press, 1991, pp. 81–111.Google Scholar
  7. Stephen M. Goldfeld, "Comments and Discussion on C. Sims' Policy Analysis with Econometric Models,"Brookings Papers on Economic Activity, I, 1982, pp. 153–157.Google Scholar
  8. C. W. J. Granger, "Investigating Causal Relations by Econometric Models and Cross-Spectral Methods,"Econometrica, 37(3), July 1969, pp. 424–438.Google Scholar
  9. David F. Hendry, "Econometric Modelling with Cointegrated Variables: An Overview,"Oxford Bulletin of Economics and Statistics, 48 (1986), pp. 201–212.Google Scholar
  10. T. C. Liu, "Underidentification, Structural Estimation and Forecasting,"Econometrica, 28 (1960), pp. 855–865.Google Scholar
  11. Robert E. Lucas, Jr., "Econometric Policy Evaluation: A Critique," in Karl Brunner and Allan H. Meltzer, editors,The Phillips Curve and Labor Markets. Amsterdam: North-Holland Publishing Co., 1976.Google Scholar
  12. H. B. Mann, and A. Wald, "On the Statistical Treatment of Linear Stochastic Difference Equations,"Econometrica, 11 (1943), pp. 173–220.Google Scholar
  13. Arthur W. Marget, "Morgenstern on the Methodology of Economic Forecasting,"Journal of Political Economy, 37, June 1929, pp. 312–339.Google Scholar
  14. Oskar Morgenstern,Wirtschaftsprognose: Eine Untersuchung ihrer Voraussetzungen und Möglichkeiten. Vienna: Springer Verlag, 1928.Google Scholar
  15. Ross, Kevin, "A Comparison of Bayesian Vector Autoregression Forecasts on the Spanish Economy: 1987–1991," Paper presented at the International Conference of the Atlantic Economic Society, Brussels, Belgium, March, 1993.Google Scholar
  16. Jeffrey D. Sachs, "Comments and Discussion on C. Sims' Policy Analysis with Econometric Models,"Brookings Papers on Economic Activity, I, 1982, pp. 157–162.Google Scholar
  17. Christopher A. Sims, "Comparison of Interwar and Postwar Business Cycles: Monetarism Reconsidered,"American Economic Review, 70, May 1980, Papers and Proceedings, pp. 250–257.Google Scholar
  18. Christopher A. Sims, "Money, Income, and Causality,"American Economic Review, 62, 1972, pp. 540–552.Google Scholar
  19. Christopher A. Sims, "Policy Analysis with Econometric Models,"Brookings Papers on Economic Activity, I (1982), pp. 107–152.Google Scholar
  20. Jan Tinbergen,An Econometric Approach to Business Cycle Problems. Paris: Hermann et Cie., 1937.Google Scholar
  21. Jan Tinbergen,Statistical Testing of Business Cycle Theories. 2 Volumes. Geneva: League of Nations, 1939.Google Scholar
  22. H. O. A. Wold, ed.,Econometric Model Building; Essays on the Causal Chain Approach. Second Printing. Amsterdam: North Holland, 1967 (c. 1964).Google Scholar
  23. Arnold Zellner, "Causality and Econometrics," in Brunner, K. and Meltzer, A. (eds.),Three Aspects of Policy and Policymaking. Amsterdam: North Holland, 1979, pp. 9–54.Google Scholar

Copyright information

© Atlantic Economic Society 1995

Authors and Affiliations

  • Brigitte H. Bechtold
    • 1
  1. 1.Central Michigan UniversityUSA

Personalised recommendations