Is there anyone who can explain as simple as possible how to go about doing this?

I have been watching different youtube videos and using google as well, however, I have not understood how to go about approaching this.

I know of the Durbin-Watson Test, of dl, du, 4, 0, 2 but it just doesn't make sense to me.

I have problems analyzing and applying this procedure for investigating whether there is evidence of autocorrelation.

I would appreciate your help.

  • $\begingroup$ @AndreSilva No. Investigating evidence of autocorrelation is a mandatory exam question and so, this is what I am focussed on. $\endgroup$ – aboabo Jul 16 '13 at 4:01
  • 2
    $\begingroup$ Are your data located in time and/or space? Autocorrelation typically refers to the similarity (correlation) between data points as the distance between points varies. So in a time series for example, the common question is whether points close in time are more similar to points more distantly separated. The Durbin-Watson statistic is constructed to test such a question - it quantifies the variance of nearby points relative to the total variance. The situation is analogous in space. $\endgroup$ – gregory_britten Jul 16 '13 at 15:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.