Time-Varying Risk Premium in Large Cross-Sectional Equity Datasets (preliminary draft)

Mardi | 2011-01-18


AbstractWe develop an econometric methodology to infer the path of risk premia from large unbalancedpanel of individual stock returns. We estimate the time-varying risk premia implied by conditional linearasset pricing models through simple two-pass cross-sectional regressions, and show consistency andasymptotic normality under increasing cross-sectional and time series dimensions. We address consistentestimation of the asymptotic variance, and testing for asset pricing restrictions. Our approach alsodelivers inference for a time-varying cost of equity. The empirical illustration on over 12,500 US stockreturns from January 1960 to December 2009 shows that conditional risk premia and cost of equitiesare large and volatile in crisis periods. They exhibit large positive and negative strays from standardunconditional estimates and follow the macroeconomic cycles. The asset pricing restrictions are rejectedfor the usual unconditional four-factor model capturing market, size, value and momentum effects butnot for its conditional version using scaled factors.aUniversity of Lugano and Swiss Finance Institute, bUniversity of Lugano, cUniversity of Genï¿oeve andSwiss Finance Institute.*Acknowledgements: We gratefully acknowledge the financial support of the Swiss National ScienceFoundation (Prodoc project PDFM11-114533).1