My understanding is that the Anderson-Hsiao adjustment, described here as an example, addresses concerns about potential or actual autocorrelation by taking a first difference of the dependent variable and adding an independent instrument which happens to be the twice lagged first difference of the dependent variable.

My question is: Why doesn't this adjustment continue to permit correlation with the once lagged first difference? If this occurs in every period, isn't that a failed correction for autocorrelation?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.