Abstract

We propose a new modeling approach for the cross section of returns. Our method, Instrumented Principal Component Analysis (IPCA), allows for latent factors and time-varying loadings by introducing observable characteristics that instrument for the unobservable dynamic loadings. If the characteristics/expected return relationship is driven by compensation for exposure to latent risk factors, IPCA will identify the corresponding latent factors. If no such factors exist, IPCA infers that the characteristic effect is compensation without risk and allocates it to an “anomaly” intercept. Studying returns and characteristics at the stock-level, we find that five IPCA factors explain the cross section of average returns significantly more accurately than existing factor models and produce characteristic-associated anomaly intercepts that are small and statistically insignificant. Furthermore, among a large collection of characteristics explored in the literature, only ten are statistically significant at the 1% level in the IPCA specification and are responsible for nearly 100% of the model’s accuracy.

Scientific Portfolio AI- Generated Summary

This paper proposes a new modeling approach for the cross section of returns called Instrumented Principal Components Analysis (IPCA). IPCA allows for latent factors and time-varying loadings by introducing observable characteristics that instrument for the unobservable dynamic loadings. By using IPCA, the authors are able to identify the corresponding latent factors that drive the characteristics/expected return relationship. If no such factors exist, IPCA infers that the characteristic effect is compensation without risk and allocates it to an “anomaly” intercept.

The authors find that IPCA produces characteristic-associated anomaly intercepts that are small and statistically insignificant, and that only eight characteristics are statistically significant in the IPCA specification and are responsible for nearly 100% of the model’s accuracy. IPCA is a powerful tool that allows the authors to incorporate latent factors and time-varying loadings into their model of risk and return, leading to more accurate predictions of asset prices and better portfolio construction.

The authors argue that many of the traditional characteristics used to explain asset pricing, such as size, value, and momentum, are actually proxies for underlying risk exposures. By incorporating these risk exposures into a unified model of risk and return, the authors are able to explain a wide range of asset pricing phenomena. They show how their model can be used to explain a number of other asset pricing phenomena, such as the value premium, the momentum premium, and the profitability premium. They also show how their model can be used to construct optimal portfolios that take into account both risk and return.

One of the key contributions of this paper is its incorporation of macroeconomic factors into the model of risk and return. The authors show how macroeconomic factors, such as inflation and growth, can be incorporated into the model to explain asset pricing phenomena. They also show how their model can be used to construct optimal portfolios that take into account both macroeconomic factors and risk and return.

Overall, “Characteristics are Covariances: A Unified Model of Risk and Return” presents a comprehensive analysis of the relationship between characteristics and covariances in the context of investment risk and return. By incorporating risk exposures into a unified model of risk and return and using IPCA to identify latent factors and time-varying loadings, the authors are able to explain a wide range of asset pricing phenomena and construct optimal portfolios that take into account both risk and return.