Hey everyone,

I need some help trying to figure out how to calculate the Durbin's Alternative Test for a set of time series econometric data using Excel.

I started with the data: y(t), x(t), x(t-1), and x(t-2). From this, I regressed the equation: y=B1+x(t)+x(t-1)+x(t-1). While doing this, I had Excel calculate the predicted y(t) and the residuals of y(t), which I will call e(t).

According to the worksheet where the data comes from, I need to calculate the alternative to the Durbin-Watson test by regressing e(t) on e(t-1) and all the regressors from the original equation.

For some reason (maybe because it is late), I cannot wrap my head around how to do the regression for this auxiliary equation. In particular, how do I calculate e(t-1) so that I can make a regression using Excel.

Anyone know? If any of this seems confusing I would be glad to clarify.