Back to basics

: can we improve the OLS CAPM?

Abstract

We provide empirical evidence of negative effects caused by endogeneity, associated with the estimation of the Capital Asset Pricing Model (CAPM) using Ordinary Least Squares (OLS), in 11 equity markets. We propose the implementation of two methods to estimate this model – Instrumental Variables (IV) and Two-Step Least Squares (2SLS). The results show that the betas obtained by using 2SLS can significantly predict future market returns a higher percentage of times than the OLS and IV approach. The 2SLS estimates also show a higher forecasting power over the future returns of the own stock.

Date of Award	10 May 2016
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Joni Kokkonen (Supervisor)