Analysis Syntax Editor - Specifications Tab

Select the Specifications tab of the General Classification and Regression Trees Syntax Editor or the General CHAID Analysis Syntax Editor to access options to "type" numbers, variable names, etc. into the syntax editor.

Click the buttons on this tab to insert the number, symbol, etc. into the program syntax at the current cursor location. For additional details, see the description of the GLM Specifications tab.

Variables

Click the Variables button to display the Variable Names to Insert dialog box. The selected variables (variable names) will be transferred to Analysis syntax edit field in the GLM Analysis Syntax Editor, at the current location of the cursor.

1, 2, ..., 0, /, . Click these buttons to insert the respective numeric character or symbol into the Analysis syntax edit field, at the current location of the cursor.

+, _+_, ..., _=_. Click these buttons to insert the respective symbols into the Analysis syntax edit field, at the current location of the cursor.

_, (, ..., ¿. Click these buttons to insert the respective symbols into the Analysis syntax edit field, at the current location of the cursor.

sigma. Click the sigma button to insert specification sigma into the Analysis syntax edit field, at the current location of the cursor; the sigma specification is an option used in conjunction with the PARAM keyword, which can be used to request the overparameterized (option overp) or sigma-restricted (option sigma) parameterization of the model.

overp. Click the overp button to insert the specification overp into the Analysis syntax edit field, at the current location of the cursor; the overp specification is an option used in conjunction with the PARAM keyword, which can be used to request the overparameterized (option overp) or sigma-restricted (option sigma) parameterization of the model.

noparam

Click the noparam button to insert the specification noparam into the Analysis syntax edit field, at the current location of the cursor; the noparam specification is an option used in conjunction with the PARAM keyword.

no. Click the no button to insert the specification no into the Analysis syntax edit field, at the current location of the cursor; the no specification is an option used in conjunction with the LACKOFFIT keyword, which can be used to request (option yes) or not request (option no) the computation of pure error to test the lack-of-fit of the overall model.

yes. Click the yes button to insert the specification yes into the Analysis syntax edit field, at the current location of the cursor; the yes specification is an option used in conjunction with the LACKOFFIT keyword, which can be used to request (option yes) or not request (option no) the computation of pure error to test the lack-of-fit of the overall model.

none. Click the none button to insert the specification none into the Analysis syntax edit field, at the current location of the cursor; the none specification is an option used in conjunction with the keywords ESTIMATE, RANDOM, SURFACE, MIXTURE, REPEATED, WDESIGN, and SAMPLE.

direct

Click the direct button to insert the specification direct into the Analysis syntax edit field, at the current location of the cursor; the direct specification is an option used in conjunction with the STOPRULE keyword, which can be used to request the FACT-style direct stopping. Splitting on the predictor variables continues until all the terminal nodes in the classification tree are "pure."

misclass

Click the misclass button to insert the specification misclass into the Analysis syntax edit field, at the current location of the cursor; the misclass specification is an option used in conjunction with the STOPRULE keyword. Specify this option to prune on the basis of misclassification error.

deviance

Click the deviance button to insert the specification deviance into the Analysis syntax edit field, at the current location of the cursor; the deviance specification is an option used in conjunction with the STOPRULE keyword. Deviance is a measure of fit that is based on the likelihood principle. This option will use the difference between the log-likelihood of the best model and the current model as a basis for pruning when the dependent variable is categorical (see Ripley, 1996). Specify this option to prune the trees on the basis of deviance.

variance

Click the variance button to insert the specification variance into the Analysis syntax edit field, at the current location of the cursor; the variance specification is an option used in conjunction with the STOPRULE keyword. One of the ways in which the size of the tree can be checked is by pruning the tree, i.e., by removing parts of trees with the aim of computing the right-sized tree. If the dependent variable is continuous (regression), the measure used is the variance of cases in a node. Specify this option to prune on the basis of variance.

gini

Click the gini button to insert the specification gini into the Analysis syntax edit field, at the current location of the cursor; the gini specification is an option used in conjunction with the GOODNESS keyword. The Gini measure (see also, Computational Formulas) is a measure of impurity of a node and can be used as a measure of goodness of fit to compute the "right-sized" tree. With priors estimated from class sizes and equal misclassification costs, the Gini measure is computed as the sum of products of all pairs of class proportions for classes present at the node. This measure reaches its maximum value when class sizes at the node are equal, and reaches a value of zero when only one class is present at a node (and hence, when the classification for the observed data is perfect). The Gini measure is the commonly preferred measure of goodness (e.g., Breiman et. al.,1984).

equal

Click the equal button to insert the specification equal into the Analysis syntax edit field, at the current location of the cursor; the equal specification is an option used in conjunction with the MISCOST and PRIOR keywords.

chisquare

Click the chisquare button to insert the specification chisquare into the Analysis syntax edit field, at the current location of the cursor; the chisquare specification is an option used in conjunction with the GOODNESS keyword. The Chi-square option is similar to the standard Chi-square value computed for the expected and observed classifications (with priors adjusted for misclassification cost).

gsquare

Click the gsquare button to insert the specification gsquare into the Analysis syntax edit field, at the current location of the cursor; the gsquare specification is an option used in conjunction with the GOODNESS keyword. The G-square option is similar to the  maximum-likelihood Chi-square (as, for example, computed in the Log-Linear module).

estimated

Click the estimated button to insert the specification estimated into the Analysis syntax edit field, at the current location of the cursor; the estimated specification is an option used in conjunction with the PRIOR keyword.

userspec

Click the userspec button to insert the specification userspec into the Analysis syntax edit field, at the current location of the cursor; the userspec specification is an option used in conjunction with the MISCOST and PRIOR keywords.