[Gretl-users] Re: Durbin-Watson statistic and Durbin's h
Am 10.11.2020 um 11:20 schrieb Alecos Papadopoulos:
Here are some reasons why
https://stats.stackexchange.com/a/148290/28746
https://stats.stackexchange.com/a/205262/28746
On 10/11/2020 10:00, Riccardo (Jack) Lucchetti wrote:
> [*] Fron an econometric point of view, I'd say that the best way to
> test for autocorrelation is using the Breusch-Godfrey LM test, but I
> digress.
In principle, of course, gretl could switch over to do the "right thing"
if lagged endogenous variables are included. It would have to be
communicated transparently somehow (not just buried in the guide
somewhere), but at some point just following the herd could also mean to
become irrelevant. (Not in the short run, though.)
cheers
sven