Title: Binary Logistic Model in Nonparametric Regression Through Spline Estimator
Authors: Dewi S Salam, Anna Islamiyati, Nirwan Ilyas
Volume: 5
Issue: 10
Pages: 50-53
Publication Date: 2021/10/28
Abstract:
Spline Logistic Regression is a modeling solution of binary categorical response data which cannot be modeled by linear regression due to violations of normality assumption. The flexibility of the spline in estimating regression curve creates a modelling approach of the regression equation that is more fitted to data than ordinary logistic regression. Spline logistic regression model parameters are estimated by using Maximum Likelihood Method with Newton Raphson's iteration. The results show that the spline logistic regression function is a non-linear estimation that depends on the number of knots of predictor and the order used in the model.