For a time series predictive regression of several excess returns, I'm looking for a way to get heteroscedasticity-consistent standard error estimates using the Wild Bootstrapping
Has this already been implemented in EViews 9 or are there any Add-Ins available?
Kind regards,
leclipse
