Coefficient of Autocorrelation

[looti]

Hi there,

was wondering if anybody could help. Econometric assignment, using eviews 3.1 (the uni's version).

I ran a regression, and include a slope dummy. Then we had to extract the residuals, and then run the resid= resid(-1) regression, to find the auto correlation coefficient (p). Then, they ask us to calculate the Durbin-Watson statistic using [ d=2(1-p) ]. Thereafter, we have to test for negative/positive autocorrelation depending on our computed d statistic.

Easy enough, but the problem lies in the fact that i dont know where to find the (p). i assumed that it would be the coefficient of resid(-1) but the d stat i get compared to the d stat provided by the regression are completely off, so im presuming that that isnt my (p).

Are the any people that could help me? im sure there is a simple way to do this, barr the long calculation method of first finding (p) (i mean, thats why we have eviews, so that it can do all of the calculations for us)

I uploaded the files as attachements (jpeg format) for referencing.

Thanks in advance for any help.

[startz]

You're looking at the wrong DW. Take a look at what you get in the second regression.