Discrete Choice Analysis Tools 2.0 provides an adaptable, efficient, and user-friendly environment for linear data classification. It’s designed with a full suite of tools built to accommodate individual model specificity, including adjustable parameter bounds, linear or nonlinear constraints, and default or user specified starting values. Newly incorporated data and parameter input procedures make model set-up and implementation intuitive.

 

Binary and count models

Binary probit

Binary logit

Negative binomial regression

Poisson regression

 

 

Multinomial logit models

Conditional logit

Nested logit

Ordered logit

Adjacent category logit

Stereotype logit

 

 

Logistic regression modelling

L2/L1 regularized classifiers

L2/L1-loss linear SVM

 

 

Accessible, storable, and exportable output

Parameter estimates

Variance-covariance matrix for coefficient estimates and marginal effects

Categorical dependent variables percentages

Data descriptions of all independent variables

Marginal effects of independent variables

Predicted counts and residuals

 

Model Selection and assessment

Full model and restricted model log-likelihoods

Chi-square statistic

Agresti’s G-squared statistic

McFadden’s Pseudo R-squared

Madalla’s Pseudo R-squared

Akaike information criterion (AIC)

Bayesian information criterion (BIC)

Likelihood ratio statistics and accompanying probability values

McKelvey and Zovcina’s psuedo R-Squared

Cragg and Uhler’s normed likelihood ratios

Count R-Squared

Adjusted count R-Squared

 

 

Examples:

Adjacent Categories Logit Model.

Binary Logit Model

Conditional Logit Model

Logistic Regression Model

Nested Logit Model

Ordered Logit Model

Stereotypical Multinomial Logit Model

 

Platform: Windows, Mac, and Linux

 

Requirements: GAUSS/GAUSS Engine/GAUSS Light v14 or higher