FORECASTING ECONOMIC TIME-SERIES USING FLEXIBLE VERSUS FIXED SPECIFICATION AND LINEAR VERSUS NONLINEAR ECONOMETRIC-MODELS

Nine macroeconomic variables are forecast in a real-time scenario usin g a variety of flexible specification, fixed specification, linear, an d nonlinear econometric models. All models are allowed to evolve throu gh time, and our analysis focuses on model selection and performance. In the context of real-time forecasts, flexible specification models ( including linear autoregressive models with exogenous variables and no nlinear artificial neural networks) appear to offer a useful and viabl e alternative to less flexible fixed specification linear models for a subset of the economic variables which we examine, particularly at fo recast horizons greater than I-step ahead. We speculate that one reaso n for this result is that the economy is evolving (rather slowly) over time. This feature cannot easily be captured by fixed specification l inear models, however, and manifests itself in the form of evolving co efficient estimates. We also provide additional evidence supporting th e claim that models which 'win' based on one model selection criterion (say a squared error measure) do not necessarily win when an alternat ive selection criterion is used (say a confusion rate measure), thus h ighlighting the importance of the particular cost function which is us ed by forecasters and 'end-users' to evaluate their models. A wide var iety of different model selection criteria and statistical tests are u sed to illustrate our findings. (C) 1997 Elsevier Science B.V.