Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/161171Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Rahardiantoro, Septian | - |
| dc.contributor.advisor | Alamudi, Aam | - |
| dc.contributor.author | Umami, Dhiffa Fatihah | - |
| dc.date.accessioned | 2025-01-31T07:09:33Z | - |
| dc.date.available | 2025-01-31T07:09:33Z | - |
| dc.date.issued | 2025 | - |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/161171 | - |
| dc.description.abstract | Analisis regresi dapat dimanfaatkan dalam pendugaan pola nonlinear. Beberapa metode yang seringkali digunakan dalam pendugaan pola nonlinear adalah regresi polinomial dan smoothing splines. Pendekatan lainnya yang dapat digunakan adalah metode generalized lasso dalam aplikasinya yaitu trend filtering dengan memanfaatkan kedekatan setiap titik waktu dengan pola polinomial pada orde tertentu. Penelitian yang akan dilakukan adalah analisis regresi nonlinear pada kasus Covid-19 di China dan Indonesia pada tahun 2020-2022. Kasus Covid-19 akan membentuk sebaran yang nonlinear. Metode regresi nonlinear yang akan digunakan, yaitu analisis regresi polinomial dan smoothing splines. Selain itu, penelitian ini juga menggunakan metode generalized lasso. Tujuan dari penelitian ini adalah mengidentifikasi pola data yang terbentuk dari ketiga metode tersebut, dimana akan ditentukan metode yang paling sesuai dalam merepresentasikan data kasus Covid-19 di China dan Indonesia dari waktu ke waktu berdasarkan nilai mean absolute percentage error (MAPE). Hasil analisis menunjukkan bahwa generalized lasso adalah metode yang paling sesuai dalam merepresentasikan pola nonlinear pada data kedua negara. Pada data kasus di China, generalized lasso dengan orde ke-4 dan nilai ?? 1SE menghasilkan nilai MAPE terkecil dan pola yang sesuai. Sementara itu, generalized lasso dengan orde ke-2 dan nilai ?? 1SE memberikan hasil terbaik untuk data Indonesia. Metode lain, seperti regresi polinomial dan smoothing splines, menunjukkan performa lebih rendah dibandingkan generalized lasso. | - |
| dc.description.abstract | Regression analysis can be utilized to estimate nonlinear patterns. Several commonly used methods for estimating nonlinear patterns include polynomial regression and smoothing splines. Another approach that can be used is the generalized lasso method, applied through trend filtering by leveraging the proximity of each time point to polynomial patterns of a certain order. This study focuses on nonlinear regression analysis of Covid-19 cases in China and Indonesia from 2020 to 2022. Covid-19 cases tend to exhibit a nonlinear distribution. The nonlinear regression methods used in this study are polynomial regression and smoothing splines. Additionally, this study employs the generalized lasso method. The purpose of this study is to identify the data patterns formed by these three methods and determine the most suitable method for representing Covid-19 case data in China and Indonesia over time based on the mean absolute percentage error (MAPE) value. The analysis results indicate that the generalized lasso method is the most suitable for representing nonlinear patterns in the data from both countries. For Covid-19 case data in China, the generalized lasso method with a 4th order and ?? 1SE yielded the smallest MAPE value and an appropriate pattern. Meanwhile, for Indonesia, the generalized lasso method with a 2nd order and ?? 1SE produced the best results. Other methods, such as polynomial regression and smoothing splines, showed lower performance compared to the generalized lasso method. | - |
| dc.description.sponsorship | null | - |
| dc.language.iso | id | - |
| dc.publisher | IPB University | id |
| dc.title | Identifikasi Pola Nonlinear Pada Kasus Covid-19 di China dan Indonesia Tahun 2020-2022. | id |
| dc.title.alternative | Identification of Nonlinear Patterns in Covid-19 Cases in China and Indonesia 2020-2022. | - |
| dc.type | Skripsi | - |
| dc.subject.keyword | generalized LASSO | id |
| dc.subject.keyword | MAPE | id |
| dc.subject.keyword | Regresi Polinomial | id |
| dc.subject.keyword | smoothing splines | id |
| dc.subject.keyword | trend filtering | id |
| Appears in Collections: | UT - Statistics and Data Sciences | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_G14190007_fc08f5c870e843558ae24aae8acfeed0.pdf | Cover | 512.65 kB | Adobe PDF | View/Open |
| fulltext_G14190007_bc3a8cb3b7fb48f9abfd9f5045ee5f4c.pdf Restricted Access | Fulltext | 1.91 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.