Klasifikasi Menggunakan Model Regresi Logistik Multinomial dan Regresi Logistik Multinomial Komponen Utama

Abstract

There are various methods that can be used in data classification. One of the classification methods is classification using multinomial logistic regression. Multinomial logistic regression is a regression model with a categorical dependent variable. In multinomial logistic regression, multicollinearity can occur. Principal component analysis is one method of dealing with multicollinearity. Multinomial logistic regression using principal component analysis is used as a comparison to ordinary multinomial logistic regression. The method used in estimating the parameter of the multinomial logistic regression model and the principal component multinomial logistic regression model is the maximum likelihood method. The significance tests and goodness-of-fit test are carried out on the models obtained. Classification is the final step after good models are obtained. Based on the data set studied, the researcher found that the accuracy of classification using the principal component multinomial logistic regression model is higher than using the ordinary multinomial logistic regression model.