Binary Logistic Regression - Storage
main topic     see also 

Stat > Regression > Binary Logistic Regression > Fit Binary Logistic Model > Storage

Stores diagnostic measures and characteristics of the estimated equation.

Dialog box items

Diagnostic Measures

Fits (event probabilities): Check to store the predicted event probabilities.

Residuals: Check to store the deviance residuals or  the Pearson residuals, depending on your selection in the Options subdialog.

Standardized residuals: Check to store the standardized deviance residuals or  the standardized Pearson residuals, depending on your selection in the Options subdialog.

Deleted residuals: Check to store the deleted deviance residuals or  the deleted Pearson residuals, depending on your selection in the Options subdialog.

Leverages: Check to store the leverages.

Cook's distance: Check to store Cook's distance.

DFITS: Check to store the DFITS.

Coefficients: Check to store the estimated coefficients for the fitted model down a column in the order that they appear in the model. See Interpreting estimated coefficients in binary logistic regression.

Design matrix: Check to store the design matrix corresponding to your model. To view the matrix in the Session window, choose Data > Display Data. This is handy for storing the design matrix for use in other Minitab commands. Minitab takes the predictors, creates the squares and cross-products, and stores all of these in a matrix. Copy the matrix into columns for use in other commands.

Variance-covariance matrix: Check to store a (d x d) matrix  (X' W X)-1, where d is the number of parameters in the model. The  (X' W X)-1 matrix is the variance-covariance matrix of the estimated coefficients.

Delta chi-square: Check to store the change in the chi-square.

Delta deviance: Check to store the change in the deviance statistic.

Delta beta (standardized): Check to store the change in the standardized estimated coefficients.

Delta beta: Check to store the change in the estimated coefficients.