A dynamic factor model framework for forecast combination

Yeung Lewis Chan; James H. Stock; Mark W. Watson

doi:10.1007/s101080050005

A dynamic factor model framework for forecast combination

Yeung Lewis Chan, James H. Stock, Mark W. Watson

Research output: Contribution to journal › Article › peer-review

64 Scopus citations

Abstract

A panel of ex-ante forecasts of a single time series is modeled as a dynamic factor model, where the conditional expectation is the single unobserved factor. When applied to out-of-sample forecasting, this leads to combination forecasts that are based on methods other than OLS. These methods perform well in a Monte Carlo experiment. These methods are evaluated empirically in a panel of simulated real-time computer-generated univariate forecasts of U.S. macroeconomic time series.

Original language	English (US)
Pages (from-to)	91-121
Number of pages	31
Journal	Spanish Economic Review
Volume	1
Issue number	2
DOIs	https://doi.org/10.1007/s101080050005
State	Published - Jul 1 1999

All Science Journal Classification (ASJC) codes

General Economics, Econometrics and Finance

Keywords

JEL classification: C32, C22
Key words: Combination forecasts, principal component regression, James-Stein estimation

Access to Document

10.1007/s101080050005

Cite this

@article{0dca4d791d6d4945b4fcc966aff66435,

title = "A dynamic factor model framework for forecast combination",

abstract = "A panel of ex-ante forecasts of a single time series is modeled as a dynamic factor model, where the conditional expectation is the single unobserved factor. When applied to out-of-sample forecasting, this leads to combination forecasts that are based on methods other than OLS. These methods perform well in a Monte Carlo experiment. These methods are evaluated empirically in a panel of simulated real-time computer-generated univariate forecasts of U.S. macroeconomic time series.",

keywords = "JEL classification: C32, C22, Key words: Combination forecasts, principal component regression, James-Stein estimation",

author = "Chan, {Yeung Lewis} and Stock, {James H.} and Watson, {Mark W.}",

note = "Publisher Copyright: {\textcopyright} 1999, Springer-Verlag Berlin Heidelberg.",

year = "1999",

month = jul,

day = "1",

doi = "10.1007/s101080050005",

language = "English (US)",

volume = "1",

pages = "91--121",

journal = "Spanish Economic Review",

issn = "1435-5469",

publisher = "Springer Verlag",

number = "2",

}

TY - JOUR

T1 - A dynamic factor model framework for forecast combination

AU - Chan, Yeung Lewis

AU - Stock, James H.

AU - Watson, Mark W.

PY - 1999/7/1

Y1 - 1999/7/1

N2 - A panel of ex-ante forecasts of a single time series is modeled as a dynamic factor model, where the conditional expectation is the single unobserved factor. When applied to out-of-sample forecasting, this leads to combination forecasts that are based on methods other than OLS. These methods perform well in a Monte Carlo experiment. These methods are evaluated empirically in a panel of simulated real-time computer-generated univariate forecasts of U.S. macroeconomic time series.

AB - A panel of ex-ante forecasts of a single time series is modeled as a dynamic factor model, where the conditional expectation is the single unobserved factor. When applied to out-of-sample forecasting, this leads to combination forecasts that are based on methods other than OLS. These methods perform well in a Monte Carlo experiment. These methods are evaluated empirically in a panel of simulated real-time computer-generated univariate forecasts of U.S. macroeconomic time series.

KW - JEL classification: C32, C22

KW - Key words: Combination forecasts, principal component regression, James-Stein estimation

UR - http://www.scopus.com/inward/record.url?scp=84952009978&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84952009978&partnerID=8YFLogxK

U2 - 10.1007/s101080050005

DO - 10.1007/s101080050005

M3 - Article

AN - SCOPUS:84952009978

SN - 1435-5469

VL - 1

SP - 91

EP - 121

JO - Spanish Economic Review

JF - Spanish Economic Review

IS - 2

ER -

A dynamic factor model framework for forecast combination

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this