What is Linear Discriminant Analysis?
Statistical modelling has several options available and Linear discriminant analysis (LDA) is one of the most popular ones for developing models. In LDA, variables are selected based on their estimated contribution to the likelihood of default. The derivation of default drivers comes from an extensive pool of qualitative features and accounting ratios.
Example of Discriminant Analysis:
Altman Z Score model is one of the classic examples of LDA. Altman’s Z-score represents the contributions (i.e. weights) of each accounting ratio to the overall score.
Why is it important?
Discriminant Analysis provides a framework for differentiating between relevant and irrelevant factors for predicting the default and assisting risk managers in building the models for default prediction.