Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/113459
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorFitrianto, Anwar-
dc.contributor.advisorRahman, La Ode Abdul-
dc.contributor.authorAtmaja, Arifah Afdila-
dc.date.accessioned2022-08-11T08:34:43Z-
dc.date.available2022-08-11T08:34:43Z-
dc.date.issued2022-
dc.identifier.urihttp://repository.ipb.ac.id/handle/123456789/113459-
dc.description.abstractGold is one of the precious metals that remain popular among investors nowadays. It provides investors a safe haven when the economy is questionable. It is crucial to provide better information about the prospect of gold prices before investing. One way to do it in the statistical realm is through forecasting using the exponential smoothing method. In order to tackle challenges that may be encountered, an additional approach called bootstrap aggregating (bagging) was proposed. This study aims to compare the performance of exponential smoothing in a condition with and without bagging. Data that were used are gold prices from January 4th, 2010 until February 27th, 2022. As there is only sign of a trend component, it can be modelled using the Holt’s linear and the additive damped trend method. Models were evaluated based on MAPE and RMSE. Based on Wilcoxon signed-rank test, adding bagging is proven to bring statistically proven difference. For the daily and weekly series, the best model is Holt’s linear smoothing with bagging. While for the monthly series, the best one is Holt’s Linear smoothing without bagging. Lastly, it is predicted that gold prices will keep increasing until reaching more than Rp1000000/g in February 2023.id
dc.description.abstractEmas merupakan salah satu logam mulia yang popular di kalangan investor. Emas dapat menjadi alternatif investasi yang aman ketika kondisi perekonomian sedang lesu. Tentu peramalan harga emas ke depan menjadi poin pertimbangan bagi investor sebelum menanamkan modalnya. Dalam ranah statististik, salah satu metode yang dapat digunakan ialah pemulusan eksponensial. Untuk mengatasi tantangan yang timbul saat proses pemodelan, bootstrap aggregating diperkenalkan sebagai pendekatan tambahan. Penelitian ini bertujuan untuk membandingkan performa pemulusan eksponensial pada kondisi dengan dan tanpa bagging serta meramalkan harga emas satu tahun ke depan. Data yang digunakan ialah data harian harga emas dari 4 Januari 2010 s.d 27 Februari 2022. Tipe pemulusan eksponensial yang digunakan ialah Holt’s linear dan the additive damped trend method. Model kemudian dievaluasi berdasarkan MAPE dan RMSE. Hasil pengujian dengan Wilcoxon signed-rank test menunjukkan bahwa penggunaan bagging sebagai pendekaan tambahan terbukti memberikan hasil peramalan yang berbeda nyata. Untuk data harian dan mingguan, model terbaik ialah pemulusan Holt’s linear dengan bagging sedangkan untuk data bulanan, model terbaik ialah pemulusan Holt’s linear tanpa bagging. Pada bulan Februari 2023, diprediksi harga emas akan melampaui Rp1000000/g.id
dc.language.isoenid
dc.publisherIPB Universityid
dc.titleGold Price Forecast through Bagging Exponential Smoothing Methodid
dc.title.alternativePeramalan Harga Emas Menggunakan Metode Bagging Pemulusan Eksponensialid
dc.typeUndergraduate Thesisid
dc.subject.keywordbaggingid
dc.subject.keywordexponential smoothingid
dc.subject.keywordgold priceid
dc.subject.keywordmapeid
dc.subject.keywordrmseid
Appears in Collections:UT - Statistics and Data Sciences

Files in This Item:
File Description SizeFormat 
Cover, Lembar Pengesahan, Prakata, Daftar Isi.pdf
  Restricted Access
Cover331.55 kBAdobe PDFView/Open
G14180026_Arifah Afdila A.pdf
  Restricted Access
Full text2.69 MBAdobe PDFView/Open
Lampiran.pdf
  Restricted Access
Lampiran749.64 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.