Peramalan Harga Saham Pt Vale Indonesia Tbk (Inco) Menggunakan Long Short Term Memory (Lstm)

Maheswara, Rakha (2025) Peramalan Harga Saham Pt Vale Indonesia Tbk (Inco) Menggunakan Long Short Term Memory (Lstm). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2043221037-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Indonesia sebagai salah satu produsen nikel terbesar di dunia memiliki peran strategis dalam rantai pasok global komoditas dan pemerintah menetapkan program hilirisasi melalui pengolahan bijih nikel menjadi produk nikel olahan, nikel menjadi komoditas yang strategis yang memiliki peran sangat penting dalam industri global sebagai bahan utama dalam pembuatan baterai kendaraan listrik, salah satu perusahaan yang bergerak pada sektor nikel yaitu PT Vale Indonesia Tbk yang berfokus pada produksi nikel. Akan tetapi harga saham bersifat fluktuatif dapat dipengaruhi faktor internal dan eksternal seperti faktor makroekonomi dan harga historis, sehingga perlu dilakukan peramalan sebagai langkah awal dalam pengambilan keputusan investasi. Peramalan saham dilakukan dengan menggunakan Long Short Term Memory (LSTM) untuk meramalkan Harga Saham INCO menggunakan data periode Januari 2007 hingga Desember 2024. Harga Saham INCO memiliki pola fluktuatif yang tajam dan sangat dipengaruhi oleh harga nikel dunia sebagai variabel utama, sedangkan variabel makroekonomi seperti Nilai Tukar, Inflasi, dan BI-Rate memiliki pengaruh yang relatif lemah. Model LSTM terbaik diperoleh dengan kombinasi variabel harga saham di masa lalu, Harga Nikel, Bi-Rate, dan Nilai Tukar dapat menghasilkan nilai MAPE sebesar 13,087% pada data train dan 6,887% pada data test. Hasil peramalan periode Januari hinggaOktober 2025 menunjukkan kecenderungan penurunan di awal periode dan kenaikan bertahap di akhir periode dengan MAPE forward testing sebesar 11,74%, sehingga model dinilai cukup baik untuk analisis tren jangka menengah.
====================================================================================================================================
Indonesia, as one of the largest nickel producers in the world, has a strategic role in the global commodity supply chain and the government has established a downstreaming program through processing nickel into processed nickel products. Nickel has become a strategic commodity that plays a crucial role in global industry as a primary material in the manufacturing of electric vehicle batteries. One company operating in the nickel sector is PT Vale Indonesia Tbk, which focuses on nickel production. However, stock prices are volatile and can be influenced by internal and external factors, such as macroeconomic factors and historical prices. Therefore, forecasting is necessary as an initial step in making investment decisions. Stock forecasting is conducted using Long Short-Term Memory (LSTM) to predict INCO's stock price using data from the period of January 2007 to December 2024. INCO's stock price exhibits a sharp, fluctuating pattern and is strongly influenced by world nickel prices as the main variable, whereas macroeconomic variables such as the Exchange Rate, Inflation, and the BI-Rate have a relatively weak influence. The best LSTM model, obtained by combining variables of past stock prices, Nickel Price, BI-Rate, and Exchange Rate, yielded a MAPE value of 13.087% on the training data and 6.887% on the testing data. The forecast results for the period of January–October 2025 indicate a downward trend at the beginning of the period and a gradual increase toward the end, with a forward testing MAPE of 11.72%. Thus, the model is considered quite good for medium-term trend analysis.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Long Short Term Memory, Makroekonomi, Peramalan, Saham Long Short Term Memory, Macroeconomics, Forecasting, Stocks
Subjects:	Q Science > Q Science (General) > Q325.78 Back propagation Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Rakha Maheswara
Date Deposited:	29 Jan 2026 04:59
Last Modified:	29 Jan 2026 04:59
URI:	http://repository.its.ac.id/id/eprint/130975

Actions (login required)

View Item