PERBANDINGAN METODE GENERALIZED CROSS VALIDATION (GCV) DAN CROSS VALIDATION (CV) DALAM MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) (STUDI KASUS: FAKTOR-FAKTOR YANG MEMPENGARUHI LAJU PERTUMBUHAN PRODUK DOMESTIK REGIONAL BRUTO PADA TAHUN 2020-2021 DI 34 PROVINSI INDONESIA)

Lathifah Siti Nur Azizah, NIM.: 19106010015 (2023) PERBANDINGAN METODE GENERALIZED CROSS VALIDATION (GCV) DAN CROSS VALIDATION (CV) DALAM MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) (STUDI KASUS: FAKTOR-FAKTOR YANG MEMPENGARUHI LAJU PERTUMBUHAN PRODUK DOMESTIK REGIONAL BRUTO PADA TAHUN 2020-2021 DI 34 PROVINSI INDONESIA). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview

Text (PERBANDINGAN METODE GENERALIZED CROSS VALIDATION (GCV) DAN CROSS VALIDATION (CV) DALAM MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) (STUDI KASUS: FAKTOR-FAKTOR YANG MEMPENGARUHI LAJU PERTUMBUHAN PRODUK DOMESTIK REGIONAL BRUTO PADA TAHUN 2020-2021 DI 3) 19106010015_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (9MB) | Preview

Text (PERBANDINGAN METODE GENERALIZED CROSS VALIDATION (GCV) DAN CROSS VALIDATION (CV) DALAM MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) (STUDI KASUS: FAKTOR-FAKTOR YANG MEMPENGARUHI LAJU PERTUMBUHAN PRODUK DOMESTIK REGIONAL BRUTO PADA TAHUN 2020-2021 DI 3) 19106010015_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (19MB) | Request a copy

Abstract

Gross Regional Domestic Product is an important indicator to determine the growth of socio-economic conditions in an area. To determine the growth rate of PDRB, the MARS model approach can be used. MARS (Multivariate Adaptive Regression Spline) is nonparametric approach whose the estimator does not require certain assumptions, does not provide information related to the regression curve but is depicted using data plots. MARS can be used on 50 n 1000 data. This study aims to compare GCV (Generalized Cross Validation) and CV (Cross Validation) methods in MARS nonparametric regression to determine the modeling form of the factors that influence Gross Regional Domestic Product in Indonesia. GCV is a method that is obtained by adding up the squared residuals that have been corrected by the squares of the factors. Meanwhile, CV is a method commonly referred to as the one-off method, which is a method that aims to minimize the sum of the squared prediction errors for the response variable. Modeling using GCV and CV produces different MSE values. GCV produces an MSE of 0.1016 with a value of R 2 = 89.69%. Then, CV produces an MSE of0.2074 with a value of R2 = 78.94%. From the result, MARS modeling with the GCV method is better used for this data than the CV method.

Share this knowledge with your friends :

Actions (login required)

View Item