Prediksi Harga Saham Menggunakan Model Geometric Brownian Motion Dengan Metode Hybrid

Azizah, Muftiyatul (2019) Prediksi Harga Saham Menggunakan Model Geometric Brownian Motion Dengan Metode Hybrid. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 06111750012003-Master_Thesis.pdf Restricted to Repository staff only Download (9MB) | Request a copy

Abstract

Saham didefinisikan sebagai tanda kepemilikan investor atas investasi mereka atau sejumlah dana yang diinvestasikan dalam suatu perusahaan. Saham merupakan instrumen yang paling dominan diperdagangkan dalam transaksi jual dan beli di bursa efek. Hasil prediksi dari pergerakan harga saham sangat penting untuk mengembangkan strategi perdagangan pasar. Prediksi harga saham dapat mengantisipasi kerugian investasi dan memberikan keuntungan optimal bagi para investor. Penelitian-penelitian tentang prediksi harga saham sudah banyak dilakukan dengan berbagai metode seperti metode geometric Brownian motion, support vector machines, jaringan saraf tiruan, dan hybrid algoritma genetika dengan jaringan saraf tiruan. Pada penelitian ini, akan dilakukan prediksi harga saham perusahaan Microsoft menggunakan metode hybrid algoritma genetika dan multilayer perceptron, serta dengan metode hybrid algoritma genetika dan geometric Brownian motion. Nilai MAPE yang dihasilkan dari hybrid algoritma genetika dan geometric Brownian motion adalah sebesar 0,0057139, sedangkan nilai MAPE yang dihasilkan oleh hybrid algoritma genetika dan multilayer perceptron adalah sebesar 0,05164. Berdasarkan nilai MAPE yang diperoleh, nilai MAPE hasil prediksi menggunakan hybrid algoritma genetika dan geometric Brownian motion lebih baik dibandingkan dengan nilai MAPE hasil prediksi menggunakan hybrid algoritma genetika dan multilayer perceptron.
================================================================================================================================
The Stock is defined as an investor ownership sign of their investment or the amount of fund invested in a company. Stock is the most instrument traded in the transaction process of stock exchange. The prediction of stock price is very important to develop a market trading strategy. The prediction of stock prices can anticipate investment losses and provide optimal benefits for investors. The researches about stock price prediction have been carried out with various methods such as geometric Brownian motion, supporting vector machines, artificial neural networks and hybrid genetic algorithms with artificial neural networks. In this research, Microsoft company stock price prediction will be carried out using hybrid genetic algorithm with multilayer perceptron method and hybrid genetic algorithm with geometric Brownian motion method. The MAPE value generated from hybrid genetic algorithm with geometric Brownian motion method is 0.0057139, while the MAPE value generated by the hybrid genetic algorithm with multilayer perceptron method is 0.05164. Based on the MAPE value obtained, the value MAPE of stock price prediction results using hybrid genetic algorithm with geometric Brownian motion method is better than the value MAPE of stock price prediction results using hybrid genetic algorithms with multilayer perceptron method.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Key-words: Stock, Geometric Brownian Motion, Multilayer Perceptron, Genetic Algorithm
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Divisions:	Faculty of Architecture, Design, and Planning > Architecture > 23201-(S1) Undergraduate Thesis
Depositing User:	AZIZAH MUFTIYATUL
Date Deposited:	23 Jul 2026 01:20
Last Modified:	23 Jul 2026 01:20
URI:	http://repository.its.ac.id/id/eprint/67751

Actions (login required)

View Item