Prediksi harga saham menggunakan moving average crossover dengan model lstm dan hybrid lstm = Stock price prediction using moving average crossover with lstm and hybrid lstm

Ranawidjaja, Ryuki (2025) Prediksi harga saham menggunakan moving average crossover dengan model lstm dan hybrid lstm = Stock price prediction using moving average crossover with lstm and hybrid lstm. Bachelor thesis, Universitas Pelita Harapan.

[thumbnail of Title] Text (Title)
Title.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (188kB)
[thumbnail of Abstract] Text (Abstract)
Abstract.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (636kB)
[thumbnail of ToC] Text (ToC)
ToC.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (666kB)
[thumbnail of Chapter1] Text (Chapter1)
Chapter1.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (557kB)
[thumbnail of Chapter2] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (918kB)
[thumbnail of Chapter3] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (696kB)
[thumbnail of Chapter4] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (3MB)
[thumbnail of Chapter5] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (550kB)
[thumbnail of Bibliography] Text (Bibliography)
Bibliography.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (563kB)
[thumbnail of Appendices] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (2MB)

Abstract

Fluktuasi harga saham menjadi tantangan bagi investor dalam investasi saham. Tujuan dari penelitian ini adalah untuk menganalisis dampak dari LSTM dengan 2 fitur input dan kombinasi LSTM dengan model CNN, ANN, dan RNN yang disertai dengan fitur keputusan jual, beli, atau tahan berdasarkan metode moving average crossover terhadap akurasi prediksi harga saham. Dilakukan hyperparameter tuning untuk mendapatkan model yang optimal untuk data harga indeks saham dari tiap negara. Penelitian ini menunjukkan bahwa model gabungan LSTM dengan CNN, ANN, atau RNN menghasilkan hasil prediksi yang lebih akurat dibandingkan dengan hasil prediksi dengan LSTM. Hal ini menunjukkan bahwa model gabungan LSTM dengan CNN, ANN, atau RNN yang disertai dengan keputusan jual atau beli memiliki potensi dalam meningkatkan akurasi prediksi harga indeks saham kedepannya yang dapat membantu dalam pengambilan keputusan investasi. LSTM-CNN menjadi model yang dominan untuk kedua kelompok negara advanced economies dan emerging and developing economies yang unggul di sebanyak 10 negara dari 20 negara, 4 dari advanced economies dengan persentase perbaikan dari LSTM yang terbaik sebesar 76% dan 6 dari emerging and developing economies dengan persentase perbaikan dari LSTM yang terbaik sebesar 88%. / Stock fluctuation pose a challenge for investors in stock investments. The aim of this research is to analyze the impact of LSTM with 2 input features and the combination of LSTM with CNN, ANN, and RNN models with buy, sell, or hold decisions based on the moving average crossover method, on stock price prediction accuracy. Hyperparameter tuning was conducted to obtain optimal models for stock index data for each country. The findings of this study indicate that the combined models of LSTM with CNN, ANN, or RNN produced more accurate prediction results compared to LSTM alone. This suggests that the integrated models of LSTM with CNN, ANN, or RNN, with sell or buy decisions, hold potential for improving the accuracy of future stock index price predictions, thereby aiding investment decision making. LSTM-CNN is the dominant model for both advanced economies and emerging and developing economies group, which perform the best in 10 countries, 4 from advanced economies with the best LSTM improvement percentage of 76% and 5 from emerging and developing economies with the best LSTM improvement percentage of 88%.
Item Type: Thesis (Bachelor)
Creators:
Creators
NIM
Email
ORCID
Ranawidjaja, Ryuki
NIM01112210032
ryuki.ranawidjaja@gmail.com
UNSPECIFIED
Contributors:
Contribution
Contributors
NIDN/NIDK
Email
Thesis advisor
Margaretha, Helena
NIDN0312057504
helena.margaretha@uph.edu
Thesis advisor
Cahyadi, Lina
NIDN0328077701
lina.cahyadi@uph.edu
Uncontrolled Keywords: lstm ; hybrid lstm ; hyperparameter tuning ; moving average crossover, prediksi saham.
Subjects: Q Science > QA Mathematics
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Depositing User: RYUKI RANAWIDJAJA
Date Deposited: 31 Jul 2025 08:42
Last Modified: 31 Jul 2025 08:42
URI: http://repository.uph.edu/id/eprint/70190

Actions (login required)

View Item
View Item