Job Market Paper:

Estimating Conditional Asset Pricing Models: Efficiency and Robustness 

This paper revisits the efficient estimation of conditional beta pricing models with traded risk factors. By applying the theory of redundant moments of Breusch, Qian, Schmidt, and Wyhowski (1999), we prove that conditional homoskedasticity of returns given the risk factors is sufficient for equilibrium pricing conditions to be redundant in the sense that they do not improve the semi-parametric efficiency bound for beta. With jointly elliptical returns and risk factors, we extend this to a necessary and sufficient condition for redundancy. Relatedly, we also prove under joint ellipticity that the optimal tuning parameter for the generalized Principal Components Analysis loadings estimator of Lettau and Pelger (2018) is the multivariate excess kurtosis coefficient of the joint distribution of the returns and risk factors. This result explains their finding that the optimal tuning parameter is zero when factors are strong and regression errors are normally distributed. A caveat for assuming a parametric model of conditional betas is the non-trivial risk of model misspecification. Motivated by the empirical work of Nagel and Singleton (2011), we proceed to evaluate constant and state-dependent parametric risk price models with an objective function that balances the level of unconditional pricing errors and the volatility of conditional pricing errors. As a benchmark for comparison, we also consider nonparametric state-dependent risk price specifications. An application to unconditional and conditional Fama and French (1993) three factor models suggests that state-dependence in risk prices delivers substantial reductions in the volatility of conditional pricing errors at a small expense to the level of unconditional errors.

Publications:

Pseudo-true SDFs in Conditional Asset Pricing Models. (with Bertille Antoine and Eric Renault), Journal of Financial Econometrics, 2018. (Online with forthcoming comments by Lars Hansen, Sydney Ludvigson, Patrick Gagliardini, Cesare Robotti and Raymond Kan)

Bernanke’s No-Arbitrage Argument Revisited: Can Open Market Operations in Real Assets Eliminate the Liquidity Trap? (with Gauti Eggertsson), in M. Woodford, editor, Monetary Policy through Asset Markets: Lessons from Unconventional Measures and Implications for an Integrated World , Santiago: Central Bank of Chile, 2016.

Master’s Thesis:

A Spatial Investigation of Urban Labor Markets. Master’s Thesis, Tufts University. 2013. (Linda Datcher Loury Award for Best Master’s Thesis)