Principal component multinomial regression is a method for modelling the relationship between a set of high-dimensional regressors and a categorical response variable with more than two categories. This method uses as covariates of the multinomial model a reduced number of principal components of the regressors. Because the principal components are based on the eigenvectors of the empirical covariance matrix, they are very sensitive to anomalous observations. Several methods for robust principal component analysis have been proposed in literature. In this study we consider ROBPCA method. The new robust approach will be applied for assessing judges' performances
A robust multinomial logit model for evaluating judges' performances
Ida Camminatiello
;
2018
Abstract
Principal component multinomial regression is a method for modelling the relationship between a set of high-dimensional regressors and a categorical response variable with more than two categories. This method uses as covariates of the multinomial model a reduced number of principal components of the regressors. Because the principal components are based on the eigenvectors of the empirical covariance matrix, they are very sensitive to anomalous observations. Several methods for robust principal component analysis have been proposed in literature. In this study we consider ROBPCA method. The new robust approach will be applied for assessing judges' performancesI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
