Prediksi tren pergerakan harga saham perbankan LQ45 dengan menggunakan metode LSTM pada Bursa Efek Indonesia

Nicky, Nicky (2022) Prediksi tren pergerakan harga saham perbankan LQ45 dengan menggunakan metode LSTM pada Bursa Efek Indonesia. Bachelor thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (82kB) | Preview
Preview	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (188kB) | Preview
Preview	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (656kB) | Preview
Preview	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (498kB) | Preview
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (312kB)
Preview	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (401kB) | Preview
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)

Abstract

In investing, no one can predict the trend of stock movement in the future. This problem makes many people afraid to invest because they want to avoid losses from the invested capital. There are various ways to predict stock movement trends in the form of candlestick analysis, and sentiment analysis and one of them is the use of Machine Learning (machine learning). In this study, machine learning using the Long Short Term Memory (LSTM) method was used to train and build a machine learning model in predicting the stock movement trend of 3 banking stocks, namely: Bank BCA, BRI, and Mandiri in the next 30 days which were processed using Python programming language. Data obtained from Yahoo! Finance from 03 January 2011 to 30 December 2021 which is downloaded and saved in Comma Spread Value (CSV) format. The development of the LSTM model in this study uses parameters of 2 LSTM layers, 200 rounds, batch size of 50, dropout of 0.2, and the use of Adam's optimization. The output of this study is the size of the data validation generated in the form of Root Mean Squared Error (RMSE) valued in 117,046 (Bank BCA), 93,380 (Bank BRI), 161,270 (Bank Mandiri), and Mean Absolute Percentage Error (MAPE) valued in 1,281 (Bank BCA), 1,935 (Bank BRI), 1997 (Bank Mandiri). In addition to RMSE and MAPE, it also calculates the optimal difference between the actual data from the test data of 20% and the predicted results from the LSTM model built./ Dalam berinvestasi tidak ada yang dapat memprediksi tren pergerakan saham pada masa depan. Masalah ini yang membuat banyak orang yang takut untuk berinvestasi karena ingin menghindari kerugian dari modal yang diinvestasikan. Terdapat berbagai macam cara untuk memprediksi tren pergerakan saham yang berupa: candlestick analysis, sentiment analysis, dan salah satunya penggunaan Machine Learning (pembelajaran mesin). Pada penelitian ini, digunakan pembelajaran mesin yang menggunakan metode Long Short Term Memory (LSTM) untuk dilatih dan dibangun model pemelajaran mesin dalam memprediksi tren pergerakan saham dari 3 saham perbankan LQ45 yaitu: Bank BCA, BRI, dan Mandiri dalam 30 hari kedepan yang diproses dengan menggunakan bahasa pemogramman Python. Data didapatkan dari Yahoo! Finance dimulai dari 03 Januari 2011 hingga 30 Desember 2021 yang diunduh dan disimpan kedalam format Comma Spread Value (CSV). Pembangunan model LSTM pada penelitian ini menggunakan parameter sebanyak 2 lapisan LSTM, 200 putaran, batch size sebanyak 50, dropout sebesar 0.2, dan penggunaan optimasi Adam. Hasil keluaran dari penelitian ini ukuran validasi data yang dihasilkan dalam bentuk Root Mean Squared Error (RMSE) senilai 117.046 (Bank BCA), 93.380 (Bank BRI), 161.270 (Bank Mandiri) dan Mean Absolute Percentage Error (MAPE) senilai 1.281 (Bank BCA), 1.935 (Bank BRI), 1.997 (Bank Mandiri). Selain RMSE dan MAPE juga dihasilkan perhitungan selisih level yang optimal antara data aktual dari data uji sebesar 20% dengan hasil prediksi model LSTM yang dibangun.

Actions (login required)

View Item