Estimator Campuran Spline Smoothing Dan Kernel Dalam Regresi Nonparametrik Multivariabel

Hidayat, Rahmat (2020) Estimator Campuran Spline Smoothing Dan Kernel Dalam Regresi Nonparametrik Multivariabel. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06211660010001-Disertation.pdf - Accepted Version
Restricted to Repository staff only

Download (3MB) | Request a copy

Abstract

Disertasi ini mengembangkan metode baru dalam mengestimasi kurva regresi nonparametrik. Metode ini menggabungkan fungsi Spline smoothing dan Kernel. Metode ini mampu menangani perbedaan pola data antara prediktor dalam regresi nonparametrik berganda. Studi teoritis difokuskan pada bentuk estimator dan pengembangan metode untuk memilih parameter penghalus dan bandwidth. Model estimator diselesaikan dengan meminimumkan Penalized Least Square (PLS). Studi simulasi dilakukan untuk menguji kinerja model yang diusulkan. Analisis empiris data tingkat pengangguran terbuka diilustrasikan untuk model yang diusulkan. Berdasarkan hasil simulasi dapat disimpulkan bahwa semakin besar ukuran sampel dan semakin kecil ukuran varians, semakin baik model yang diperoleh. Dalam analisis data riil, estimator campuran Spline smoothing dan Kernel mampu memodelkan tingkat pengangguran terbuka dengan GCV = 0,0011818 dan R2 = 88,58%. ============================================================================================================== This dissertation develops new method in estimating nonparametric regression curve. This method combines the smoothing Spline and Kernel functions. This method is able to handle differences in data patterns between predictors in multiple nonparametric regression. Theoretical study is focused on the form of estimator and development a method to selecting the smoothing parameters and bandwidth. Estimation of the estimator is completed by minimizing Penalized Least Square (PLS). Simulation studies are conducted to examine the performance of the proposed model. An empirical analysis of the unemployment rate data is illustrated for the proposed methodology. Based on the simulation results can be concluded that the larger the sample size and the smaller the size of the variance, the better the model obtained. In the real data analysis, this model is able to model the unemployment rate with GCV = 0.0011818 and R2 = 88.58%.

Item Type: Thesis (Doctoral)
Additional Information: RDSt 519.536 Hid e-1
Uncontrolled Keywords: nonparametric regression, smoothing Spline, Kernel, Penalized Least Squares (PLS), regresi nonparametrik, Spline smoothing, Kernel, Penalized Least Squares (PLS)
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Science and Data Analytics (SCIENTICS)
Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49001-(S3) PhD Thesis
Depositing User: RAHMAT HIDAYAT
Date Deposited: 06 Aug 2020 06:50
Last Modified: 10 Aug 2020 04:25
URI: https://repository.its.ac.id/id/eprint/76217

Actions (login required)

View Item View Item