Nominal Logistic Regression - Storage
main topic     see also  

Stat > Regression > Nominal Logistic Regression > Storage

Stores characteristics of the estimated equation, event probabilities, and aggregated data.

Dialog box items

Characteristics of Estimated Equation

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

Standard error of coefficients: Check to store the standard errors of the estimated coefficients down a column in the order that they appear in the model.

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.

Log-likelihood for last iteration: Check to store the last log-likelihood in a constant.

Event probabilities and/or aggregated data

Number of trials: Check to store the number of trials for each factor/covariate pattern.

For the remaining two storage items

Enter the number of events: Enter the number of levels (distinct values) of the response variable.

Event probabilities: Check to store the predicted event probabilities, one probability for each distinct value of the response.

Number of occurrences: Check to store the number of occurrences for the jth factor/covariate pattern.