Forecasting with macroeconometric models: A report from the trenches

Arbeitspapier

Econometric models are a widely used and powerful tool in macroeconomic analysis and forecasting. Admittedly, their acceptance by the scientific community has had some hard times during the seventies and eighties: a general decline in the reputation of macroeconomics, the Lucas critique, and failures of the model community to make their often opaque practice transparent had left their marks. Closer looks at the criticism, however, revealed its limited relevance, and the „new/old macroeconomic consensus“ (Blinder, Zarnowitz) of the early 1990s seems to have restored much of the lost credibility. A lack of transparency, however, still diminishes acceptance and credibility of the results, at least within the academic community3. The apparent deficit in model transparency has a number of causes, a major one being the fact that the literature on the practice of macroeconometric forecasting is still sparse (cf. e.g., Klein, Young 1980, pp. 75ff., Adams 1986, pp. 106ff.) and, given the new technical opportunities for model and forecast analysis, it is also somewhat outdated. One explanation for this is that economic deliberations in the model industry back such disregard (Daub 1987, pp. 73ff.) and it may still take some time until the industry realises that transparency is the models' biggest asset. This paper describes in detail the production of a macroeconometric forecast, complementing and enlarging on two earlier papers on the subject (Heilemann 1985, 1990). The model used is the business cycle model of the Rheinisch-Westfälisches Institut für Wirtschaftsforschung (RWI) which has been applied for forecasting since 1978.

Angaben zum Objekt

• Sprache: Englisch

Verweise und Beziehungen

• Erschienen in: Series: Technical Report ; No. 1999,47

Klassifikation und Themen

• Thema: Konjunkturprognose ; Ökonometrisches Makromodell ; Prognoseverfahren ; Zeitreihenanalyse ; Deutschland

Beteiligte, Orts- und Zeitangaben

• Ereignis: Geistige Schöpfung
• (wer): Heilemann, Ullrich
• Ereignis: Veröffentlichung
• (wer): Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen
• (wo): Dortmund
• (wann): 1999

Weitere Informationen

• Handle: http://hdl.handle.net/10419/77184
• Letzte Aktualisierung: 10.03.2025, 11:45 MEZ