What is discriminant function analysis used for?

What is discriminant function analysis used for?

• A. A test to determine if variances are equal across groups.
• B. A linear combination of variables created such that the differences between group means on the transformed variable are maximized.
• C. A method to estimate the probability of an event.
• D. A technique to produce confidence intervals for means.

Answer: (B)

Discriminant function analysis builds a linear combination of the predictor variables to create a discriminant score that makes the predefined groups as different as possible on that score. By maximizing the differences between group means on the transformed variable while considering within-group variability, it produces a function that best separates the groups. This discriminant function can then be used to classify new observations into those groups and to see which variables contribute most to the separation. It’s not about testing equal variances, not about directly estimating event probabilities (that’s more in the realm of logistic approaches), and not about forming confidence intervals for means.

Discriminant function analysis builds a linear combination of the predictor variables to create a discriminant score that makes the predefined groups as different as possible on that score. By maximizing the differences between group means on the transformed variable while considering within-group variability, it produces a function that best separates the groups. This discriminant function can then be used to classify new observations into those groups and to see which variables contribute most to the separation. It’s not about testing equal variances, not about directly estimating event probabilities (that’s more in the realm of logistic approaches), and not about forming confidence intervals for means.