A brief review of statistical models for prediction of categorical data is presented, with emphasis on the binary type. Several methods have been adopted to build predictive models for binary and other types of categorical data and response variables. The focus here is on generalized linear models and generalized additive models, widely applied in problems in Ecology, when the goal is to fit a model to data of presence/absence type or any other categorical response. The estimation methods used for generalized linear models and generalized additive models as well its statistical properties are discussed. Some examples in ecology are addressed.
|Title of host publication
|Proceedings of the International Conference on Numerical Analysis and Applied Mathematics 2014, ICNAAM 2014
|Theodore E. Simos, Theodore E. Simos, Theodore E. Simos, Charalambos Tsitouras
|American Institute of Physics Inc.
|Published - Mar 10 2015
|International Conference on Numerical Analysis and Applied Mathematics 2014, ICNAAM 2014 - Rhodes, Greece
Duration: Sep 22 2014 → Sep 28 2014
|AIP Conference Proceedings
|International Conference on Numerical Analysis and Applied Mathematics 2014, ICNAAM 2014
|9/22/14 → 9/28/14
All Science Journal Classification (ASJC) codes
- General Physics and Astronomy