14.03.2014 Views

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

Modeling and Multivariate Methods - SAS

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Chapter 13 Recursively Partitioning Data 327<br />

Partition Method<br />

K Fold Crossvalidation shows a Crossvalidation report, giving fit statistics for both the training <strong>and</strong><br />

folded sets. For more information on validation, see “Validation” on page 335.<br />

ROC Curve<br />

Lift Curve<br />

is described in the section “ROC Curve” on page 337. This is for categorical responses only.<br />

is described in the section “Lift Curves” on page 339. This is for categorical responses only.<br />

Show Fit Details shows several measures of fit <strong>and</strong> a confusion matrix. The confusion matrix is a<br />

two-way classification of actual <strong>and</strong> predicted response. This is for categorical responses only.<br />

Entropy RSquare compares the log-likelihoods from the fitted model <strong>and</strong> the constant probability<br />

model.<br />

Generalized RSquare is a generalization of the Rsquare measure that simplifies to the regular Rsquare<br />

for continuous normal responses. It is similar to the Entropy RSquare, but instead of using the<br />

log-likelihood, it uses the 2/n root of the likelihood. It is scaled to have a maximum of 1. The value is 1<br />

for a perfect model, <strong>and</strong> 0 for a model no better than a constant model.<br />

Mean -Log p is the average of -log(p), where p is the fitted probability associated with the event that<br />

occurred.<br />

RMSE is the root mean square error, where the differences are between the response <strong>and</strong> p (the fitted<br />

probability for the event that actually occurred).<br />

Mean Abs Dev is the average of the absolute values of the differences between the response <strong>and</strong> p (the<br />

fitted probability for the event that actually occurred).<br />

Misclassification Rate is the rate for which the response category with the highest fitted probability is<br />

not the observed category.<br />

For Entropy RSquare <strong>and</strong> Generalized RSquare, values closer to 1 indicate a better fit. For Mean -Log p,<br />

RMSE, Mean Abs Dev, <strong>and</strong> Misclassification Rate, smaller values indicate a better fit.<br />

Save Columns<br />

is a submenu for saving model <strong>and</strong> tree results, <strong>and</strong> creating <strong>SAS</strong> code.<br />

Save Residuals saves the residual values from the model to the data table.<br />

Save Predicteds saves the predicted values from the model to the data table.<br />

Save Leaf Numbers saves the leaf numbers of the tree to a column in the data table.<br />

Save Leaf Labels saves leaf labels of the tree to the data table. The labels document each branch that<br />

the row would trace along the tree, with each branch separated by “&”. An example label could be<br />

“size(Small,Medium)&size(Small)”. However, JMP does not include redundant information in the form<br />

of category labels that are repeated. When a category label for a leaf references an inclusive list of<br />

categories in a higher tree node, JMP places a caret (‘^”) where the tree node with redundant labels<br />

occurs. Therefore, “size(Small,Medium)&size(Small)” is presented as ^&size(Small).<br />

Save Prediction Formula saves the prediction formula to a column in the data table. The formula is<br />

made up of nested conditional clauses that describe the tree structure. The column includes a Response<br />

Probability property.<br />

Save Tolerant Prediction Formula saves a formula that predicts even when there are missing values.<br />

The column includes a Response Probability property.<br />

Save Leaf Number Formula saves a column containing a formula in the data table that computes the<br />

leaf number.

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

Saved successfully!

Ooh no, something went wrong!