Discriminant Analysis
overview     how to     example     data     see also 

Stat > Multivariate > Discriminant Analysis

Use discriminant analysis to classify observations into two or more groups if you have a sample with known groups. Discriminant analysis can also used to investigate how variables contribute to group separation.

Minitab offers both linear and quadratic discriminant analysis. With linear discriminant analysis, all groups are assumed to have the same covariance matrix. Quadratic discrimination does not make this assumption but its properties are not as well understood.

In the case of classifying new observations into one of two categories, logistic regression may be superior to discriminant analysis [3], [11].

Dialog box items

Groups: Choose the column containing the group codes. There may be up to 20 groups.

Predictors: Choose the column(s) containing the measurement variables or predictors.

Discriminant Function

Linear: Choose to perform linear discriminant analysis. All groups are assumed to have the same covariance matrix.

Quadratic: Choose to perform quadratic discriminant analysis. No assumption is made about the covariance matrix; its properties are not as well understood.

Use cross validation: Check to perform the discrimination using cross-validation. This technique is used to compensate for an optimistic apparent error rate.

Storage

Linear discriminant function: Enter storage columns for the coefficients from the linear discriminant function, using one column for each group. The constant is stored at the top of each column.

Fits: Check to store the fitted values. The fitted value for an observation is the group into which it is classified.

Fits from cross validation: Check to store the fitted values if discrimination was done using cross-validation.

<Options>