Multinomial logistic regression

Multinomial logistic regression is useful for situations in which you want to be able to classify subjects based on values of a set of predictor variables. This type of regression is similar to logistic regression, but it is more general because the dependent variable is not restricted to two categories.

Example
In order to market films more effectively, movie studios want to predict what type of film a moviegoer is likely to see. By performing a Multinomial Logistic Regression, the studio can determine the strength of influence a person's age, gender, and dating status has upon the type of film they prefer. The studio can then slant the advertising campaign of a particular movie toward a group of people likely to go see it.
Statistics
Iteration history, parameter coefficients, asymptotic covariance and correlation matrices, likelihood-ratio tests for model and partial effects, –2 log-likelihood. Pearson and deviance chi-square goodness of fit. Cox and Snell, Nagelkerke, and McFadden R 2. Classification: observed versus predicted frequencies by response category. Crosstabulation: observed and predicted frequencies (with residuals) and proportions by covariate pattern and response category.
Methods
A multinomial logit model is fit for the full factorial model or a user-specified model. Parameter estimation is performed through an iterative maximum-likelihood algorithm.

Data considerations

Data
The dependent variable should be categorical. Independent variables can be factors or covariates. In general, factors should be categorical variables and covariates should be continuous variables.
Assumptions
It is assumed that the odds ratio of any two categories are independent of all other response categories. For example, if a new product is introduced to a market, this assumption states that the market shares of all other products are affected proportionally equally. Also, given a covariate pattern, the responses are assumed to be independent multinomial variables.

Obtaining a Multinomial logistic regression

This feature requires Custom Tables and Advanced Statistics.

  1. From the menus choose:

    Analyze > Association and prediction > Multinomial logistic regression

  2. Click Select variable under the Dependent variable section and select a single dependent variables. Click OK after selecting the variable.
  3. Optionally, click the link next to the selected variable (for example, Last Category (Ascending)) to open the Define Reference Category dialog. The dialog provides options for specifying reference category settings. Verify or change the default settings and then click OK.
  4. Optionally, click Select variables under the Independent variables > Factors section to select categorical, independent factor variables that might have an influence on the dependent variable.
  5. Optionally, click Select variables under the Independent variables > Covariates section to select continuous, independent variables that might have an influence on the dependent variable.
  6. You can optionally expand the Additional settings menu and select the following options.
    • Click Model to specify the effects to be analyzed.
    • Click Convergence criteria to specify criteria parameters such as iterations, convergence values, delta, and singularity tolerance.
    • Click Statistics to select which statistics to include in the analysis.
    • Click Options specify constant, stepwise probability, classification, iteration, memory, and missing value settings.
    • Click Save to dataset to add casewise post-estimation statistics to the dataset as new variables.
    • Click Model export to export parameter estimates and their covariances to an external XML file.
    • Click Bootstrap for deriving robust estimates of standard errors and confidence intervals for estimates such as the mean, median, proportion, odds ratio, correlation coefficient or regression coefficient.

This procedure pastes NOMREG command syntax.