PREDIKSI HARGA SAHAM PT BRI Tbk MENGGUNAKAN DEEP LEARNING LONG SHORT TERM MEMORY DENGAN OPTIMASI STOCHASTIC GRADIENT DESCENT

KUSUMA, HAYU FIRDAUS (2022) PREDIKSI HARGA SAHAM PT BRI Tbk MENGGUNAKAN DEEP LEARNING LONG SHORT TERM MEMORY DENGAN OPTIMASI STOCHASTIC GRADIENT DESCENT. Sarjana / Sarjana Terapan (S1/D4) thesis, Universitas Muhammadiyah Semarang.

[img]
Preview
Text
HALAMAN JUDUL.pdf

Download (439kB) | Preview
[img]
Preview
Text
ABSTRACK.pdf

Download (503kB) | Preview
[img]
Preview
Text
BAB 1.pdf

Download (734kB) | Preview
[img]
Preview
Text
BAB II.pdf

Download (1MB) | Preview
[img] Text
BAB III.pdf
Restricted to Repository staff only

Download (754kB) | Request a copy
[img] Text
BAB IV.pdf
Restricted to Repository staff only

Download (779kB) | Request a copy
[img]
Preview
Text
BAB V.pdf

Download (425kB) | Preview
[img]
Preview
Text
DAFTAR PUSTAKA.pdf

Download (518kB) | Preview
[img] Text
LAMPIRAN.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy
[img] Text
SKRIPSI FULL TEKS.pdf
Restricted to Repository staff only

Download (2MB) | Request a copy
Official URL: http://repository.unimus.ac.id

Abstract

ABSTRAK Kusuma Hayu Firdaus, 2022. Prediksi Harga Saham PT BRI Tbk menggunakan Deep learning Long Short Term Memory dengan Optimasi Stochastic Gradient Descen. Skripsi, Program Studi Statistika Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Muhammadiyah Semarang. Pembimbing: I. Indah Manfaati Nur, S.Si.M.Si, II. Prizka Rismawati Arum, M.Stat. Prediksi harga saham sangat dibutuhkan oleh para investor untuk menjadi bahan pertimbangan sebelum penanaman modal. Dalam penelitian ini akan dilakukan prediksi harga saham PT Bank Rakyat Indonesia Tbk menggunakan Deep Learning Long Short Term Memory (LSTM) pada harga saham peutupan selama 5 tahun dengan periode 1 Januari 2016 sampai dengan 31 Desember 2021. LSTM banyak dipilih oleh peneliti untuk prediksi berbasis waktu atau time series karena dikenal dalam mengatasi gradient yang menghilang atau dapat mengatasi masalah data dalam jangka waktu yang panjang. Untuk meningkatkan akurasi dalam pembuatan model prediksi, penelitian ini menggunakan algoritma optimasi Stochastic Gradient Descen (SGD). Hasil penelitian ini menunjukan bahwa model terbaik prediksi LSTM dengan error MSE sebesar 5,17 dan skenario jumlah neuron sebanyak 30 dan epoch 250 dengan learning rate 0,1. Dari hasil perhitungan prediksi harga saham penutupan BRI memiliki tingkat akurasi 98,7% dimana nilai tersebut menunjukan bahwa hasil prediksi sangat baik. Kata kunci: LSTM, Prediksi Harga Saham, Stochastic Gradient Descen (SGD).   ABSTRACT Kusuma Hayu Firdaus, 2022. PT BRI Tbk Stock Price Prediction using Deep learning Long Short Term Memory with Stochastic Gradient Descen Optimization. Undergraduate Thesis, Statisics Study Program Faculty of Mathematics and Natural Sciences, University of Muhammadiyah Semarang. Supervisor: I. Indah Manfaati Nur, S.Si.M.Si, II. Prizka Rismawati Arum, M.Stat. Stock price predictions are needed by investors to be taken into consideration before investing. In this study, stock price predictions of PT Bank Rakyat Indonesia Tbk will be carried out using Deep Learning Long Short Term Memory (LSTM) on closing stock prices for 5 years with a period of January 1, 2016 to December 31, 2021. LSTM is widely chosen by researchers for time-based predictions or time series because it is known for dealing with disappearing gradients or for solving data problems over a long period of time. To improve accuracy in making prediction models, this study used the Stochastic Gradient Descend (SGD) optimization algorithm. The results of this study indicate that the best model predicts LSTM with an MSE error of 5.17 and a scenario of 30 neurons and 250 epochs with a learning rate of 0.1. From the results of the calculation of the closing stock price prediction, BRI has an accuracy rate of 98.7% where this value indicates that the prediction results are very good. Keywords : LSTM, Stock Price Prediction, Stochastic Gradient Descen (SGD).

Item Type: Thesis (Sarjana / Sarjana Terapan (S1/D4) )
Call Number: 014/Statistika/IX/2022
Subjects: L Education > Statistics
Divisions: Faculty of Science and Mathematics > S1 Statistics
Depositing User: perpus unimus
URI: http://repository.unimus.ac.id/id/eprint/6018

Actions (login required)

View Item View Item