In Microsoft Excel 2000, you can use the Analysis ToolPak add-in to perform a
variety of data analysis functions, including linear regression analysis.
Steps to Reproduce the Problem
The following steps demonstrate the problem described earlier.
1. | In Microsoft Excel, create a new workbook. In Sheet1, enter the following data:
A1: 1 B1: 9
A2: 2 B2: 8
A3: 3 B3: 7
A4: 4 B4: 6
|
2. | On the Tools menu, click Data Analysis. In the Data Analysis dialog box, click Regression and then click OK.
|
3. | In the Regression dialog box, in the Input Y Range box, type $A$1:$A$4.
|
4. | In Input X Range box, type $B$1:$B$4 .
|
5. | Under Output options, click Output Range, and then type $D$1 in the box to the right of the option button.
|
6. | Click OK.
The Summary Output table appears in the worksheet, starting in cell D1. If you scroll to the cell range I16:L18, you see the following data:
Lower 95% Upper 95% Lower 95.0% Upper 95.0%
10 10 10 10
-1 -1 -1 -1
The first two columns display the lower and upper 95%, and the last two columns also display the lower and upper 95.0%. This is true because the Confidence Level check box in the Regression dialog box was not selected.
|
7. | On the Tools menu, click Data Analysis. In the Data Analysis dialog box, click Regression and then click OK.
|
8. | Click to select the Confidence Level check box. Type 85 in the box to the right of the check box.
|
9. | Click OK, and then click OK again to overwrite the existing table.
|
The following data appears in cells I16:L18:
Lower 95% Upper 95% Lower 85.0% Upper 85.0%
10 10 10 10
-1 -1 -1 -1
The first two columns still display the lower and upper 95%; the last two
columns now display the lower and upper 85.0%.