eprintid: 64966
rev_number: 10
eprint_status: archive
userid: 12460
dir: disk0/00/06/49/66
datestamp: 2024-04-24 04:39:07
lastmod: 2024-04-24 04:39:07
status_changed: 2024-04-24 04:39:07
type: thesis
metadata_visibility: show
contact_email: muh.khabib@uin-suka.ac.id
creators_name: Linggar Utami, NIM.: 20106010003
title: OPTIMISASI ANALISIS REGRESI LASSO PADA KASUS KRIMINALITAS DI INDONESIA TAHUN 2021 DENGAN ALGORITMA LARS MENGGUNAKAN AIC
ispublished: pub
subjects: Matematika
divisions: jur_mat
full_text_status: restricted
keywords: Multikolinearitas, LASSO, LARS, AIC, Kriminalitas
note: Pembimbing: Sri Utami zuliana, S.Si., M.Sc., Ph.D
abstract: Multicollinearity is a common problem in regression analysis, where several
independent variables have a high correlation between their variables. Identification
of multicollinearity uses the Value of Variance Inflation Factor (VIF) or an
examination of the correlation coefficient between variables in a regression model.
Multicollinearity problems have significant impacts, including an increase in variance
and the incidence of bias, which can ultimately lead to overfitting and lower
coefficient significance. Therefore, addressing multicollinearity is a critical aspect
to ensure the stability and accuracy of the regression model developed. To overcome
these problems, the LASSO method (Least Absolute Shrinkage and Selection
Operator) is used. LASSO regression uses a method of shrinking the coefficients of
independent variables effectively directing the coefficients of variables that have a
high correlation to approach or reach zero values. The search for LASSO regression
coefficients is carried out using the Least Absolute Regression (LARS) algorithm. In
this study, the selection of optimal lambda in LASSO regression uses the Akaike’s
Information Criterion (AIC). This research is applied to crime data in Indonesia in
2021. The conclusion obtained based on the research that has been done is that
there is one of the seven independent variables included in the LASSO regression
model, namely the total population in Indonesia (X2) where this variable affects the
crime rate in Indonesia in 2021.
date: 2024-03-08
date_type: published
pages: 93
institution: UIN SUNAN KALIJAGA YOGYAKARTA
department: FAKULTAS SAINS DAN TEKNOLOGI
thesis_type: skripsi
thesis_name: other
citation:   Linggar Utami, NIM.: 20106010003  (2024) OPTIMISASI ANALISIS REGRESI LASSO PADA KASUS KRIMINALITAS DI INDONESIA TAHUN 2021 DENGAN ALGORITMA LARS MENGGUNAKAN AIC.  Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.   
document_url: https://digilib.uin-suka.ac.id/id/eprint/64966/1/20106010003_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf
document_url: https://digilib.uin-suka.ac.id/id/eprint/64966/2/20106010003_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf