Suppose that a response variable y is related to four predictor variables, 1 x , 2 x , 3 x , and 4 x , so that k54.

1. Enter the observed values of y and each of the k54 predictor variables into the first

(k11) columns of an Excel spreadsheet. (note: In order for the multiple regression analysis to work properly, there must be a column of values for each independent predictor variable xi in your model, and the x columns must be adjacent to each other.)

2. Use Data ➤ Data Analysis ➤ Regression to generate the Dialog box, highlighting or typing in the cell ranges for the xi and y values and check “Labels” if necessary.

3. If you click “Confidence Level,” Excel will calculate confidence intervals for the regression estimates, 0 b , 1 b , 2 b , 3 b , and 4 b . Enter a cell location for the Output Range and click OK to generate the regression output.

note: MS Excel does not provide options for estimation and prediction. Also, the diagnostic plots which can be generated in Excel are not the same plots as we have discussed in Section 13.2 and will not be discussed in this section.