Prediksi Harga Saham Jangka Pendek di Indonesia Menggunakan Metode Gaussian Process Regression

Gultom, Elnora Oktaviyani (2021) Prediksi Harga Saham Jangka Pendek di Indonesia Menggunakan Metode Gaussian Process Regression. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06111740000093-Undergraduate_Thesis.pdf] Text
06111740000093-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (5MB) | Request a copy
[thumbnail of 06111740000093-Undergraduate_Thesis.pdf] Text
06111740000093-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (5MB) | Request a copy

Abstract

Prediksi harga saham umumnya dilakukan secara jangka panjang, sedangkan harga saham tiap waktu mengalami perubahan yang signifikan. Tujuan dari Tugas Akhir ini untuk memprediksi harga saham jangka pendek dengan membangkitkan model Gaussian Process Regression menggunakan beberapa kernel yang berbeda. Model dengan menggunakan kernel Rational Quadratic dan RBF memiliki nilai rata-rata RMSE terkecil dibandingkan kedua kernel lainnya. Prediksi harga saham berdasarkan waktu dengan menggunakan kernel tersebut diperoleh prediksi satu minggu kedepan menghasilkan nilai EVS sebesar 0.99871. Dari hasil penelitian pada data historis harga saham 01 Desember 2019 sampai 25 Februari 2021, prediksi harga saham minggu berikutnya dihasilkan bahwa perusahaan PT Gudang Garam Tbk memiliki nilai jual yang paling tinggi dan PT United Tractors Tbk memiliki nilai beli lebih murah. Sedangkan perusahaan pada sektor Consumer Non Cyclical memiliki rata-rata nilai jual dengan return yang tertinggi dan sektor Industrial memiliki nilai rata-rata harga beli saham dalam jumlah lebih banyak yang tertinggi.
================================================================================================
Stock price forecasts are often made on a long-term basis, despite the fact that stock prices fluctuate significantly from time to time. The goal of this study is to forecast short-term stock values by creating the Gaussian Process Regression model using a variety of kernels. When compared to the other two kernels, the model utilizing the Rational Quadratic kernel with RBF has the smallest average RMSE value. Prediction of stock prices based on time using the kernel yields an EVS value of 0.99871 for the coming week. Based on research on historical stock price data from December 1, 2019 to February 25, 2021, the forecast of stock prices the next week revealed that PT Gudang Garam Tbk had the greatest selling value and PT United Tractors Tbk had the lowest purchasing value. Meanwhile, the Consumer Non-Cyclical sector has the greatest average selling value with the largest return, while the Industrial sector has the highest average share buying price.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Prediksi harga saham, Gaussian Process Regression, Pola Jangka Pendek, Kernel, Stock Proce Prediction, Gaussian Process Regression, Short Term, Kernel
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QA Mathematics > QA246.8 Gaussian
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User: Elnora Oktaviyani Gultom
Date Deposited: 27 Aug 2021 06:26
Last Modified: 27 Aug 2021 06:26
URI: http://repository.its.ac.id/id/eprint/90123

Actions (login required)

View Item View Item