Estimasi Model Campuran Spline Truncated Dan Kernel Dalam Regresi Nonparametrik Multivariabel

., Rismal (2016) Estimasi Model Campuran Spline Truncated Dan Kernel Dalam Regresi Nonparametrik Multivariabel. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 1314201009-Master_Thesis.pdf]
Preview
Text
1314201009-Master_Thesis.pdf - Accepted Version

Download (1MB) | Preview

Abstract

data berpasangan dengan variabel prediktor dan variabel respon yi diasumsikan mengikuti pola data yang tidak diketahui sedemikian sehingga dapat didekati oleh model regresi nonparametrik campuran Komponen didekati dengan fungsi spline aditif dengan prediktor sebanyak p sementara gk (t) didekati dengan fungsi kernel aditif dengan prediktor sebanyak . Error diasumsikan berdistribusi normal dengan mean nol dan varians konstan. Dengan menggunakan metode Maximum Likelihood Estimation (MLE) diperoleh estimator , dan juga adalah matriks yang bergantung pada knot dan bandwidth. Walaupun estimator yang diperoleh merupakan estimator bias, Namun estimator tersebut merupakan kelas estimator linear dalam observasi y . Hasil teoritis diterapkan pada data tingkat pengangguran terbuka tahun 2013 di Jawa Barat. Model yang dihasilkan memberikan nilai R2 sebesar 78.05%.
======================================================================================================
Given a pair of data with predictors and response variables yiare assumed to follow unknown function such that their dependence can be approximated by a mixture nonparametric regression model is approximated with additive spline functions with p-number of predictors whereas is approximated with additive kernel functions with q-number of predictors. The error is assumed normally distributed with mean zero and constant variance. By means of Maximum Likelihood Estimation (MLE) method, it was obtained , as well as are the matrices depending on knots and bandwidths. The estimators obtained was empirically applied on open unemployment rate in West Java in 2013. The resulted model yielded R2 value 78.05%

Item Type: Thesis (Masters)
Additional Information: RTSt 519.536 Ris e
Uncontrolled Keywords: Regresi Nonparametrik Campuran, Spline, Kernel, MLE, Tingkat Pengangguran Terbuka.
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions: Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis
Depositing User: Yeni Anita Gonti
Date Deposited: 17 Apr 2020 07:13
Last Modified: 17 Apr 2020 08:09
URI: http://repository.its.ac.id/id/eprint/75803

Actions (login required)

View Item View Item