When to use Linear Discriminant Analysis or Logistic Regression

Question

The Wikipedia article on Logistic Regression says:

Logistic regression is an alternative to Fisher's 1936 method, linear discriminant analysis. If the assumptions of linear discriminant analysis hold, application of Bayes' rule to reverse the conditioning results in the logistic model, so if linear discriminant assumptions are true, logistic regression assumptions must hold. The converse is not true, so the logistic model has fewer assumptions than discriminant analysis and makes no assumption on the distribution of the independent variables.

Could someone help me to understand what the assumptions of linear discriminant are, an example of where they hold, and how application of Bayes' rule results in the Logistic Model.

Would it be correct to say that Logistic Regression is always the preferred choice? Are there conditions where Linear Discriminant Analysis is to be preferred?

Answer 1

This question was asked and answered at the Cross-Validate SE. The answer is a few years old but it is still useful. If you want you can click here to find their answer.