 Hi, this is Dr. Don. I want to show you how to use phstat and Excel to get the Durban Watson statistic. A number of students have sent me questions saying that they're getting errors in their Durban Watson output. So I think I know what's going on. Let me show you how not to get the Durban Watson correct. Let's take our data here. We're going to go to phstat, regression, simple linear. Our output, our response variable or wide variable. I'm going to select USA sales, highlight that range. My predictor variable is going to be my month. And I'm going to highlight that range all the way down. I make sure they match, which they do. My first sale does have a label. I want the 95% confidence interval. This is the way the dialog box pops up as the default. The regression statistics table is checked. That gives you some information, but nothing else is checked. So I'm suspecting that students are going down here and just selecting Durban Watson. Some of them have selected the scatter plot and they'll get the confidence interval. But the thing they do not select is the ANOVA, and that's the part that kills them. Well, let's go ahead and click OK, and we get some output here. We get the regression statistics table. Our beta one, our beta zero coefficients are there and some other information, or our square information, but we don't have the ANOVA, as I mentioned. Jump over here to the Durban Watson. There's the error messages that students seem to be getting. If you click on one of those, you can see where the linkages are. It goes back to the residuals page. Click on residuals. There's more of those error messages. And if we click there, we can see that's just subtracting on this tab. But if I click on that one, it goes to the compute. It doesn't give me the references, but it says something is missing there. If you go back to compute, again, what's missing is the ANOVA. But let me show you. I'm going to go back here to my data. I'm going to go to regression. Simple linear again. I'm going to select my sales for USA. Very quickly there. I'm going to select my Y variable, X variable. I'm going to click from the bottom this time. You can go either way, of course. Make sure they match, which they do for sales. This time I'm going to click the ANOVA and I'm going to get my Durban Watson statistic and then click the residuals table, residuals plot, just to make sure. And now we've got the complete compute table there. I go to my Durban Watson table. I've got it complete. If I go to, let's see again, that's the residuals 2. There's residuals 2. This is complete. And if we click there, you can see the references B18 and B17 on compute 2. If I go over here, it is this B17, the beta 0, the intercept and beta 1, the slope. That is what it's missing. So that's what you need to do to get your Durban Watson statistic to come out properly.