Robust Inference for Consumption-Based Asset Pricing
Frank Kleibergen and Zhaoguo Zhan
Journal of Finance Vol. 75, No. 1 (February 2020), pp. 507-550
Overview The reliability of traditional asset-pricing tests depends on (i) the correlations between asset returns and factors; and (ii) the time-series sample size T compared to the number of assets N. However, for macro-risk factors, like consumption growth, they often cannot be trusted. We extend the Gibbons-Ross-Shanken statistic to identify risk premiums and construct their 95-percent confidence sets. These sets are wide or unbounded when T and N are close, but when T exceeds N by a considerable margin, average returns are not fully spanned by betas. Our approach enables meaningful empirical inferences.
10 | Journal Publications - Financial Times Top 50 Journals
Made with FlippingBook - Online catalogs