Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Durbin Watson and Panel data

Options
  • 29-08-2013 10:37pm
    #1
    mickyboy
    Registered Users Posts: 39 ✭


    Hi boards.

    I keep getting low DW stats for my panel data regressions. When I fix it using AR(1), Cochrane Orcutt, first differencing etc my coefficients become insignificant while DW approaches 2.

    Just wondering if I should really be worried about DW in panel data.

    Thanks


Comments

  • Options
    #2
    Ostrom
    Registered Users Posts: 3,483 ✭✭✭Ostrom


    Absolutely - autocorrelation can inflate standard errors, bias parameter estimates, and lead to erroneous inference. Did you try using alternative specifications? Autoregressive error models are one route, but using alternative lag combinations can soak up the serial correlation, and give you a little more information on the long-term effects.


  • Options
    #3
    andrew
    Moderators, Science, Health & Environment Moderators, Society & Culture Moderators Posts: 3,368 Mod ✭✭✭✭andrew


    Moved to the 'researcher' forum, since the marginal return on an extra day in the Economics forum was likely very low at this stage


  • Options
    #4
    freeze4real
    Registered Users Posts: 2,540 ✭✭✭freeze4real


    andrew wrote: »
    Moved to the 'researcher' forum, since the marginal return on an extra day in the Economics forum was likely very low at this stage

    Haha I see what you did there.

    I like it.


Advertisement