14.03.2014 Views

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Chapter 3 Fitting St<strong>and</strong>ard Least Squares Models 89<br />

Effect Screening<br />

The last three Effect Screening options in the report’s red triangle menu show a plot <strong>and</strong> the Parameter<br />

Estimate Population report. The Parameter Population report finds the correlation of the estimates <strong>and</strong> tells<br />

you whether the estimates are uncorrelated <strong>and</strong> have equal variances.<br />

• If the estimates are correlated, a normalizing transformation can be applied make them uncorrelated <strong>and</strong><br />

have equal variances.<br />

• If an estimate of error variance is unavailable, the relative st<strong>and</strong>ard error for estimates is calculated by<br />

setting the error variance to one.<br />

• If the estimates are uncorrelated <strong>and</strong> have equal variances, then the following notes appear under the<br />

Effect Screening report title:<br />

– The parameter estimates have equal variances.<br />

– The parameter estimates are not correlated.<br />

If the estimates are correlated <strong>and</strong>/or have unequal variances, then each of these two notes can appear as list<br />

items. The selected list item shows that JMP has transformed the estimates.<br />

To undo both transformations, click these list items.<br />

The transformation to uncorrelate the estimates is the same as that used to calculate sequential sums of<br />

squares. The estimates measure the additional contribution of the variable after all previous variables have<br />

been entered into the model.<br />

Example of Data with Equal Variances <strong>and</strong> Uncorrelated Estimates<br />

1. Open the Reactor.jmp sample data table.<br />

2. Select Analyze > Fit Model.<br />

3. Select Y <strong>and</strong> click Y.<br />

4. Make sure that the Degree box has a 2 in it.<br />

5. Select Ct, A, T, <strong>and</strong> Cn <strong>and</strong> click Macros > Factorial to Degree.<br />

6. Click Run.<br />

7. From the red triangle menu next to Response Y, select Effect Screening > Normal Plot.<br />

8. Open the Parameter Estimate Population report.<br />

Figure 3.28 Parameter Estimate Population Report for Reactor.jmp

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!