The poor predictive performance of asset pricing models

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

This paper examines time-series forecast errors of expected returns from conditional and unconditional asset pricing models for portfolio and individual firm equity returns. A new result that increases predictive precision concerning model specification and forecasting is introduced. Conditional versions of the models generally produce higher mean squared errors than unconditional versions for step ahead prediction. This holds for individual firm data when the instruments are firm specific. Mean square forecast error decompositions indicate that the asset pricing models produce relatively unbiased predictions, but the variance is severe enough to ruin the step ahead predictive ability beyond that of a constant benchmark.

Original languageEnglish (US)
Pages (from-to)355-380
Number of pages26
JournalJournal of Financial and Quantitative Analysis
Volume43
Issue number2
StatePublished - Jun 1 2008

Fingerprint

Prediction
Asset pricing models
Forecast error
Mean squared error
Expected returns
Equity returns
Benchmark
Ruin
Model specification
Decomposition
Predictive ability

All Science Journal Classification (ASJC) codes

  • Accounting
  • Finance
  • Economics and Econometrics

Cite this

@article{6d212c2a5c4c4ea9bff71e220671c26f,
title = "The poor predictive performance of asset pricing models",
abstract = "This paper examines time-series forecast errors of expected returns from conditional and unconditional asset pricing models for portfolio and individual firm equity returns. A new result that increases predictive precision concerning model specification and forecasting is introduced. Conditional versions of the models generally produce higher mean squared errors than unconditional versions for step ahead prediction. This holds for individual firm data when the instruments are firm specific. Mean square forecast error decompositions indicate that the asset pricing models produce relatively unbiased predictions, but the variance is severe enough to ruin the step ahead predictive ability beyond that of a constant benchmark.",
author = "Simin, {Timothy T.}",
year = "2008",
month = "6",
day = "1",
language = "English (US)",
volume = "43",
pages = "355--380",
journal = "Journal of Financial and Quantitative Analysis",
issn = "0022-1090",
publisher = "Cambridge University Press",
number = "2",

}

The poor predictive performance of asset pricing models. / Simin, Timothy T.

In: Journal of Financial and Quantitative Analysis, Vol. 43, No. 2, 01.06.2008, p. 355-380.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The poor predictive performance of asset pricing models

AU - Simin, Timothy T.

PY - 2008/6/1

Y1 - 2008/6/1

N2 - This paper examines time-series forecast errors of expected returns from conditional and unconditional asset pricing models for portfolio and individual firm equity returns. A new result that increases predictive precision concerning model specification and forecasting is introduced. Conditional versions of the models generally produce higher mean squared errors than unconditional versions for step ahead prediction. This holds for individual firm data when the instruments are firm specific. Mean square forecast error decompositions indicate that the asset pricing models produce relatively unbiased predictions, but the variance is severe enough to ruin the step ahead predictive ability beyond that of a constant benchmark.

AB - This paper examines time-series forecast errors of expected returns from conditional and unconditional asset pricing models for portfolio and individual firm equity returns. A new result that increases predictive precision concerning model specification and forecasting is introduced. Conditional versions of the models generally produce higher mean squared errors than unconditional versions for step ahead prediction. This holds for individual firm data when the instruments are firm specific. Mean square forecast error decompositions indicate that the asset pricing models produce relatively unbiased predictions, but the variance is severe enough to ruin the step ahead predictive ability beyond that of a constant benchmark.

UR - http://www.scopus.com/inward/record.url?scp=46849120008&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46849120008&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:46849120008

VL - 43

SP - 355

EP - 380

JO - Journal of Financial and Quantitative Analysis

JF - Journal of Financial and Quantitative Analysis

SN - 0022-1090

IS - 2

ER -