Extracting Business Cycle Fluctuations: What Do Time Series Filters Really Do?

36 Pages Posted: 26 Jul 2007

See all articles by Arturo Estrella

Arturo Estrella

Rensselaer Polytechnic Institute

Date Written: June 2007

Abstract

Various methods are available to extract the "business cycle component" of a given time series variable. These methods may be derived as solutions to frequency extraction or signal extraction problems and differ in both their handling of trends and noise and their assumptions about the ideal time-series properties of a business cycle component. The filters are frequently illustrated by application to white noise, but applications to other processes may have very different and possibly unintended effects. This paper examines several frequently used filters as they apply to a range of dynamic process specifications and derives some guidelines for the use of such techniques.

Keywords: frequency domain, spectral analysis, signal extraction

JEL Classification: C22, E32

Suggested Citation

Estrella, Arturo, Extracting Business Cycle Fluctuations: What Do Time Series Filters Really Do? (June 2007). FRB of New York Staff Report No. 289, Available at SSRN: https://ssrn.com/abstract=1003178 or http://dx.doi.org/10.2139/ssrn.1003178

Arturo Estrella (Contact Author)

Rensselaer Polytechnic Institute ( email )

110 8th Street
Troy, NY 12180
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
250
Abstract Views
1,349
Rank
222,467
PlumX Metrics