PERAMALAN HARGA SAHAM SYARIAH DENGAN MENGGUNAKAN HYBRID CONVOLUTIONAL NEURAL NETWORK – LONG SHORT TERM MEMORY

Wanda Alifia, - (2022) PERAMALAN HARGA SAHAM SYARIAH DENGAN MENGGUNAKAN HYBRID CONVOLUTIONAL NEURAL NETWORK – LONG SHORT TERM MEMORY. S1 thesis, Universitas Pendidikan Indonesia.

Abstract

Saham merupakan salah satu alat investasi yang saat ini menarik banyak
perhatian dari masyarakat. Saham dapat memberikan keuntungan yang tinggi,
namun dapat pula menimbulkan kerugian besar dalam waktu yang singkat karena
sifatnya yang fluktuatif. Pergerakan nilai saham dipengaruhi oleh beberapa faktor,
salah satunya dipengaruhi oleh nilai historis dari saham tersebut. Oleh karena itu,
peramalan harga saham sudah banyak dilakukan dan menjadi penelitian yang
penting karena dapat dijadikan bahan pertimbangan investor dalam melakukan
investasi saham. Beberapa model telah diterapkan untuk dapat meramalkan
fluktuasi harga saham, namun jika hanya menggunakan sebuah model tunggal,
model tersebut tentunya memiliki keterbatasan. Sehingga pada penelitian ini,
dibangun model hybrid Convolutional Neural Network (CNN) – Long Short Term
Memory (LSTM) dengan cara menggabungkan algoritma terbaik untuk dapat
memanfaatkan kelebihan yang dimiliki oleh masing-masing algoritma. Hasil
penelitian menunjukkan bahwa model hybrid CNN-LSTM menghasilkan tingkat
akurasi yang baik dalam peramalan harga saham.
Kata kunci: Saham, investasi saham, Convolutional Neural Network, Long
Short Term Memory, hybrid CNN-LSTM

Currently, stocks have attracted much attention among the public as a
means of investment. Although stocks can offer high returns, they may also result
in large losses in a short period of time due to its volatility. Stock value movements
are influenced by several factors, including the historical value of the stock.
Therefore, stock market forecasting is widely practiced and becomes an important
area of research as it can be a consideration for investors in making stock
investments. A number of models have been applied to predict stock price
fluctuations, but when using only a single model, the model certainly has
limitations. Therefore, in this study, a hybrid Convolutional Neural Network (CNN)
– Long-Short Term Memory (LSTM) model was developed by combining the best
algorithms to maximize the advantages of each algorithm. The results show that the
CNN-LSTM hybrid model yields a high accuracy level in stock price forecasting.
Keyword: Stocks, stock investment, Convolutional Neural Network, Long�Short Term Memory, hybrid CNN-LSTM.

	Text S_MAT_1804316_Title.pdf Download (359kB)
	Text S_MAT_1804316_Chapter1.pdf Download (225kB)
	Text S_MAT_1804316_Chapter2.pdf Restricted to Staf Perpustakaan Download (482kB)
	Text S_MAT_1804316_Chapter3.pdf Download (304kB)
	Text S_MAT_1804316_Chapter4.pdf Restricted to Staf Perpustakaan Download (731kB)
	Text S_MAT_1804316_Chapter5.pdf Download (103kB)
	Text S_MAT_1804316_Appendix.pdf Restricted to Staf Perpustakaan Download (1MB)

Item Type:	Thesis (S1)
Additional Information:	ID Sinta Dosen Pembimbing 1 : Dewi Rachmatin, S.Si., M.Si. (5975775) ID Sinta Dosen Pembimbing 2 : Fitriani Agustina, S.Si., M.Si. (5981275)
Uncontrolled Keywords:	Saham, investasi saham, Convolutional Neural Network, Long Short Term Memory, hybrid CNN-LSTM
Subjects:	L Education > L Education (General) Q Science > QA Mathematics
Divisions:	Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Matematika - S1 > Program Studi Matematika (non kependidikan)
Depositing User:	Wanda Alifia
Date Deposited:	15 Nov 2022 07:55
Last Modified:	15 Nov 2022 07:55
URI:	http://repository.upi.edu/id/eprint/82692

Actions (login required)

View Item