Factor Alignment Problems and Quantitative Portfolio Management

Abstract

Quantitative equity portfolio management has evolved into an interdisciplinary activity that draws expertise from the fields of finance, statistics, econometrics, accounting, and optimization. Each one of these streams is a mature discipline in itself, having its own body of knowledge, and operates under assumptions that are usually well accepted within the respective communities. But when concepts from these diverse fields are applied in a common setting, there is bound to be friction among various assumptions that get further magnified due to the use of an optimizer. In this article, Ceria, Saxena, and Stubbs focus on the interaction of three key elements that are part of the quantitative portfolio management process, namely, the expected returns model, the risk model, and the constraints that are used to formulate the portfolio construction problem. They generally refer to the issues caused by this interaction as factor alignment problems. The authors present a detailed investigation of these alignment problems, survey some of their common sources, analyze and document their effects on the ex post performance of optimized portfolios, and conclude with a practical and effective remedy in the form of augmented risk models.

TOPICS: Equity portfolio management, statistical methods, portfolio construction