 We're going to use the Excel Data Analysis Toolpack to conduct a regression analysis of this data. I'm going to go to Data, Data Analysis, open up the dialog box, look down to I find Regression, select it, click OK, bring up the second dialog box. We need to input the Y range, I'm going to clear those out. Again, this is the Y, the response variable. This is a place that a lot of students make a mistake. Make sure you're selecting your Y variable data. Notice I only selected one cell with a label, you don't want to select two cells, that will mess you up. And we're going to click Label to indicate that we've got labels. I'm going to clear out the X, and again I'm going to select just the single label drag down to select that range. Here's a point, place to double check to make sure that you have an equal number of Y values and X values, you can do that just by comparing the cell references there. Again, make sure that the labels are selected, I'm going to click on Confidence and leave it at 95%. The output I want to put into this worksheet, so I'm going to delete that and select that cell. And I am going to go ahead and leave the line fit plot selected, that will give us a scatter plot with the regression lines plotted for us. These other options you can use, and you may need some other problems. I'm going to click OK, and we get our standard regression output. Over here we've got our line plot, which is kind of small, but you can drag that around to expand it, and then you can format it to meet your desires. Remember by selecting one data point with a left click and then right click, we can add a trend line. I'm linear again, we want to display the equation in R square, click that off, and so we've got our equation in R square on our scatter plot. Going back to our output here, I'm going to expand these cells so you can see them a little better. Up here we have the regression output. This is the correlation coefficient that R, about 91%, which is high. This is the ANOVA, and that tells us whether or not the regression is significant. I'm going to change that so you can make it a little more understandable, get away from that scientific notation, do the same thing down here on these p-values. That tells me this F-test that the overall ANOVA is significant, and that means that there is a slope that is different from zero. At the bottom we've got our intercept, 20,512, which is the same as we see over here, and we've got our coefficient a negative 9.51 as price increases, demand drops down. We can see here that the p-values on both of those are less than our alpha of 0.05, which means that they're both significant, statistically significant. We also have the confidence interval around the intercept and around the slope down at the bottom. I'm going to scroll here. We have our observations, 20 observations, and the predicted demand based on this regression equation, and then the residuals are the difference between the actual value at that observation and the predicted value of that observation.