Toda, Alexis Akira and Walsh, Kieran James, Fat Tails and Euler Equations: A Monte Carlo Study (September 11, 2014). Available  for download at SSRN: http://ssrn.com/abstract=2495151

“The Euler equation estimation of heterogeneous-agent consumption-based asset pricing models is challenging because the cross-sectional consumption distribution may have fat tails and is measured with error. We generate asset returns and consumption data from an incomplete markets dynamic general equilibrium model that exhibits a known power law in consumption and is analytically tractable. Monte Carlo experiments suggest that the standard GMM estimation of the relative risk aversion coefficient is biased upwards and susceptible to type II errors. The “age cohort” estimation method suggested in Toda and Walsh (2014) appears to be consistent and to mitigate type II errors.”