Sri Pradnyapramitha Arolla Ramadhani Oktafianto, Trianingsih Eni Lestari
Multiple linear regression analysis is a statistical technique to investigate and model the relationship between a response variable (Y) and multiple predictors (X). One of the assumptions that must be met from multiple regression is the absence of multicollinearity. Multicollinearity is a condition that occurs in multiple regression analysis when there is a strong correlation or relationship between two or more predictor variables. Multicollinearity in this study can be overcome by LASSO regression. LASSO regression reduces the estimated coefficient to zero and selects predictor variables to produce the best model. The LASSO regression model can be estimated using the LARS and GLMNet algorithms because both algorithms are computationally fast in estimating the LASSO regression coefficient. This research uses HDI 2020 data from 29 districts/cities in Papua. The results obtained indicate that the LASSO regression using the LARS algorithm is better than GLMNet in overcoming the multicollinearity case, with the R-squared value obtained being 99.5%. Several factors significantly influence HDI, namely Expenditures Per Capita, Life Expectancy, Expected Length of School, and Average Length of Schooling. © 2024 Author(s).
Department of Mathematics, Universitas Negeri Malang, Jl. Semarang 5, Malang, 65145, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock