QUADRATIC LOGISTIC REGRESSION MODEL AND PRINCIPAL CURVES APPLIED TO DISCRIMINANT ANALYSIS: COMPARISON OF METHODS

Favorite this paper
How to cite this paper?
Details
  • Presentation type: Trabalho completo (oral)
  • Track: 9. EST&MP – Estatística e Modelos Probabilísticos
  • Keywords: Logistic Regression; Principal Curves; Discriminant Analysis;
  • 1 Universidade Tecnológica Federal do Paraná

QUADRATIC LOGISTIC REGRESSION MODEL AND PRINCIPAL CURVES APPLIED TO DISCRIMINANT ANALYSIS: COMPARISON OF METHODS

Inácio Andruski-Guimarães

Universidade Tecnológica Federal do Paraná

Abstract

Studies on logistic regression usually consider the model with linear discriminant functions, but there are situations where quadratic discriminant functions work better. However, the quadratic logistic regression model involves the estimation of a great number of parameters. Another problem is that the model may not work when the data set has no overlapping. To deal with this issue we propose an extension of the hidden logistic model, whose advantage is the existence of estimators, even when there is no overlap. Principal curves, that pass through the middle of a p-dimensional data set, providing a summary of the data in ℝp, have been used for discriminant analysis. We propose an algorithm which extracts principal curves from the Fisher discriminant matrix, using the k-segments algorithm. This algorithm generates a polygonal line used to classify an observation in the class whose polygonal line has the shortest Euclidean distance by orthogonal projection.

Share your ideas or questions with the authors!

Did you know that the greatest stimulus in scientific and cultural development is curiosity? Leave your questions or suggestions to the author!

Sign in to interact

Have a question or suggestion? Share your feedback with the authors!