Research Info

Home \Comparison of Subspace ...
Title
Comparison of Subspace Dimension Reduction Methods in Logistic Regression
Type Article
Keywords
Dimension reduction, Likelihood acquired direction, Sliced average variance estimation, Sliced inverse regression.
Abstract
Regression models are very useful in describing and predicting real world phenomena. When the number of predictors in regression models is high, data analysis is difficult. Dimension reduction has become one of the most important issues in regression analysis because of its importance in dealing with problems with high-dimensional data. In this paper, the methods of diminishing the dimension of variables, which include the estimation of central subspace based on the inverse regression, the likelihood acquisition method and principal component analysis are considered. Using a real data associated with the dental problems the Logistic regression is fitted and the correct classification of the data computed. The simulation study is presented to compare the sufficient dimension reduction methods with each other.
Researchers saeed Heydari (First researcher) , Mahmoud Afshari (Second researcher) , Saeid Tahmasebi (Third researcher) , Morad Alizadeh (Fourth researcher)