I'm estimating a model with a lagged dependent variable and fixed effects for panel units. I understand that this can result in Nickell Bias but I didn't think it would be a big problem because there are more time periods than panel units. A reviewer has asked that we demonstrate that Nickell Bias is not a problem. How might I go about this? I understand the Sargan-Hansen test is potentially useful, but I've only seen it applied to generalized method of moments (GMM) estimators. Is it possible to run something like Sargan-Hansen in the context of a regular LDV model?




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.