I was listening to a talk where the presenter was talking about using regularized estimation approaches in a VAR(1) model $$X_t = Gamma X_{t-1} + epsilon_t, quad epsilon_t sim mathcal{N}(0,Omega).$$ The claim was put forth that deviations from normality assumptions, e.g. that in this setting are less of a problem because of the regularization approach. Is this true in general? I have performed a reference search in Google as well as Google Scholar but could not really find anyhting. Is there any general theory about regularization with respect to deviations from normality?
↧