EViews.com

Does EViews offer a Lagrange multiplier (LM) or F-statistic analog for two stage least squares regression (2SLS)? (For example, as in Wooldridge, 1990).

I would like to use EViews to test for serial correlation in for 2SLS regression in two cases: i) with and ii) without lagged dependent variables on the right hand side of the equation. However, from what I understand, the Durbin-Watson and Breusch-Godfrey tests are only consistent under OLS.

Thanks

The reported Breusch-Pagan includes the adjustments to the statistics as described in Wooldridge (1990). The latter is included in our list of references, though we do not discuss it in the text of the manual.