Discussion Paper

Modeling trends

Abstract: Models of low-frequency behavior of time series may have strongly conflicting substantive implications while fitting the data nearly equally well. We should develop methods which display the resulting uncertainty rather than adopt modeling conventions which hide it. One step toward this goal may be to consider overparameterized stationary ARMA models.

Bibliographic Information

Provider: Federal Reserve Bank of Minneapolis

Part of Series: Discussion Paper / Institute for Empirical Macroeconomics

Publication Date: 1989

Number: 22