Autocorrelation in panel data

[AngieL]

Hi everyone,

Does anyone know how to estimate the autocorrelation for panel data?

Thank you in advance,
Angie

[ahmed slamah]

According to my knowledge, The Durbin-Watson statistic is used to test for autocorrelation using E-views program. The Durbin-Watson test yields a result between 0 and 4, with 0 indicating positive autocorrelation and 4 indicating negative autocorrelation. A result close to 2 provides sufficient confidence that no autocorrelation is present.

[AngieL]

Thank you very much, but I am not sure if I can use the DW for the panel data.

[bayu_sutikno]

Angiel, Why you are not sure can use Durbin Watson statistics to test autocorrelation ?? If any the other method what's name its ? I very need this topic. Thank you.

[EViews Glenn]

The manual has an example of using the Wooldridge framework for testing against an AR(1) in a panel setting.

http://www.eviews.com/help/helpintro.ht ... es.html%23