This paper develops a semiautoregression approach to estimate factors of the arbitrage pricing theory that has the advantage of providing a simple asymptotic variance-covariance matrix for the factor estimates, which makes it easy to adjust for measurement errors. Using the extracted factors, the author confirms the finding that the arbitrage pricing theory describes asset returns slightly better than the capital asset pricing model, although there is still some mispricing in the arbitrage pricing theory model. The author finds that not only are the factors priced by the market, but the factor premiums move over time in relation to business cycle variables.

[In this article we break assets' betas with common factors into components attributable to news about future cash flows, real interest rates, and excess returns. To achieve this decomposition, we use a vector autoregressive time-series model and an approximate log-linear present value relation. The betas of industry and size portfolios with the market are largely attributed to changing expected returns. Betas with inflation and industrial production reflect opposing cash flow and expected return effects. We also show how asset pricing theory restricts the expected excess return components of betas.]

This paper uses an autoregressive approach to test a multi-factor model with time-varying risk premiums. A quasi-differencing approach is used to eliminate the unobservable factors in the model. It is found that the model is capable of capturing the “size effect” and the “dividend yield effect, ” but is incapable of explaining the “book-to-market effect” and the “earnings-price ratio effect.” Thus, it is concluded that a constant-beta multi-factor model will not be able to explain the cross-sectional variation in expected returns.

This paper developes a semiautoregression (SAR) approach to estimate factors of the arbitrage pricing theory (APT) that has the advantage of providing a simple asymptotic variance‐covariance matrix for the factor estimates, which makes it easy to adjust for measurement errors. Using the extracted factors, I confirm the finding that the APT describes asset returns slightly better than the CAPM, although there is still some mispricing in the APT model. I find that not only are the factors “priced” by the market, but the factor premiums move over time in relation to business cycle variables.

In this article we break assets’ betas with common factors into components attributable to news about future cash flows, real interest rates, and excess returns. To achieve this decomposition, we use a vector autoregressive time-series model and an approximate log-linear present value relation. The betas of industry and size portfolios with the market are largely attributed to changing expected returns. Betas with inflation and industrial production reflect opposing cash flow and expected return effects. We also show how asset pricing theory restricts the expected excess return components of betas.

In this article we break assets’ betas with common factors into components attributable to news about future cash flows, real interest rates, and excess returns. To achieve this decomposition, we use a vector autoregressive time-series model and an approximate log-linear present value relation. The betas of industry and size portfolios with the market are largely attributed to changing expected returns. Betas with inflation and industrial production reflect opposing cash flow and expected return effects. We also show how asset pricing theory restricts the expected excess return components of betas.