Stochastic Volatility, Long Run Risks, and Aggregate Stock Market Fluctuations

BIS Working Papers  |  No 323  | 
26 October 2010


What are the main drivers of fluctuations in the aggregate US stock market? In this paper, we attempt to resolve the long-lasting debate surrounding this question by designing and solving a consumption-based asset pricing model which incorporates stochastic volatility, long-run risks in consumption and dividends, and Epstein-Zin preferences. Utilizing Bayesian MCMC techniques, we estimate the model by fitting it to US data on the level of the aggregate US stock market, the short-term real risk-free interest rate, real consumption growth, and real dividend growth. Our results indicate that, over short and medium horizons, fluctuations in the level of the aggregate US stock market are mainly driven by changes in expected excess returns. Conversely, low frequency movements in the aggregate stock market are primarily driven by changes in the expected long-run growth rate of real dividends.

JEL Classification: G12, E21, E44, C11

Keywords: asset pricing, stochastic volatility, long-run risks, Bayesian MCMC Methods