Cookies?
Library Header Image
LSE Theses Online London School of Economics web site

Essays in empirical asset pricing

Bryzgalova, Svetlana (2015) Essays in empirical asset pricing. PhD thesis, The London School of Economics and Political Science (LSE).

[img]
Preview
Text - Submitted Version
Download (4MB) | Preview

Abstract

In this thesis, I study asset pricing models of stock and bond returns, and therole of macroeconomic factors in explaining and forecasting their dynamics. The first chapter is devoted to the identification and measurement of risk premia in the cross-section of stocks, when some of the risk factors are only weakly related to asset returns and, as a result, spurious inference problems are likely to arise. I develop a new estimator for cross-sectional asset pricing models that, simultaneously, provides model diagnostic and parameter estimates. This novel approach removes the impact of spurious factors and restores consistency and asymptotic normality of the parameter estimates. Empirically, I identify both robust factors and those that instead suffer from severe identification problems that render the standard assessment of their pricing performance unreliable (e.g. consumption growth, human capital proxies and others). The second chapter extends the shrinkage-based estimation approach to the class of affine factor models of the term structure of interest rates, where many macroeconomic factors are known to improve the yield forecasts, while at the same time being unspanned by the cross-section of bond returns. In the last chapter (with Christian Julliard), we propose a simple macro model for the co-pricing of stocks and bonds. We show that aggregate consumption growth reacts slowly, but significantly, to bond and stock return innovations. As a consequence, slow consumption adjustment (SCA) risk, measured by the reaction of consumption growth cumulated over many quarters following a return, can explain most of the cross-sectional variation of expected bond and stock returns. Moreover, SCA shocks explain about a quarter of the time series variation of consumption growth, a large part of the time series variation of stock returns, and a significant (but small) fraction of the time series variation of bond returns, and have substantial predictive power for future consumption growth.

Item Type: Thesis (PhD)
Additional Information: © 2015 Svetlana Bryzgalova
Library of Congress subject classification: H Social Sciences > HB Economic Theory
Sets: Departments > Economics
Supervisor: Julliard, Christian and Robinson, Peter
URI: http://etheses.lse.ac.uk/id/eprint/3204

Actions (login required)

Record administration - authorised staff only Record administration - authorised staff only

Downloads

Downloads per month over past year

View more statistics