The Cross Section of Expected Returns with MIDAS Betas

Abstract

This paper explores the cross-sectional variation of expected returns for a large cross section of industry and size/book-to-market portfolios. We employ mixed data sampling (MIDAS) to estimate a portfolio’s conditional beta with the market and with alternative risk factors and innovations to well-known macroeconomic variables. The market risk premium is positive and significant, and the result is robust to alternative asset pricing specifications and model misspecification. However, the traditional 2-pass ordinary least squares (OLS) cross-sectional regressions produce an estimate of the market risk premium that is negative, and significantly different from 0. Using alternative procedures, we compare both beta estimators. We conclude that beta estimates under MIDAS present lower mean absolute forecasting errors and generate better out-of-sample performance of the optimized portfolios relative to OLS betas.

How Wealthy are the Rich?

“…In contrast, the rich list is a carefully selected sample aimed at covering the super-rich, and one can therefore expect that undersampling is not an issue. On the contrary, the manager magazin staff relies on public records for their compilation of the 14 There exists an ongoing debate on whether this relationship can be established empirically (French et al, 1987;Campbell, 1987;Nelson, 1991;Campbell and Hentschel, 1992;Harvey, 2001;Goyal and Santa-Clara, 2003;Brandt and Kang, 2004;Ghysels et al, 2005;Bali and Peng, 2006;Andersen et al, 2006;Guo and Whitelaw, 2006;Lundblad, 2007;Bali, 2008;Gonzales et al, 2012). While most of the studies find at least weak support for the risk-return trade-off for various time frames and markets, the debate seems to have now shifted to the precise functional form of this relationship-as opposed to the linear one implied by the CAPM.…”

Predicting Systematic Risk With Macroeconomic and Financial Variables

“…Other studies such as Bollerslev, Patton, and Quaedvlieg (2016) and Hollstein, Prokopczuk, and Simen (2020) show that one can improve on these estimates/ forecasts using even higher frequency intraday data. Another strand of this large literature (e.g., Gonzalez, Nave, and Rubio 2012;Cenesizoglu, Liu, and Reeves 2016;Cenesizoglu and Reeves 2018) shows that one can use both high-and low-frequency data jointly to estimate and forecast betas. Given the theoretical justification as well as the overwhelming empirical evidence in favor of using high-frequency data to estimate and forecast betas, we focus on predicting the quarterly time series of realized betas estimated using daily data.…”

Asset Pricing Using Block-Cholesky GARCH and Time-Varying Betas

“…To overcome these issues González, Nave, and Rubio (2012) propose a linear beta pricing model with time-varying risk exposures based on a mixed frequency approach. The conditional betas are estimated using a kernel-based weighted realized variance estimator, which exploits highfrequency information.…”