LQ45 STOCK PRICE PREDICTION USING MONTE CARLO SIMULATION AND LONG-SHORT TERM MEMORY

• Main Author: Leonardo Aritonang, Martin
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/78400
• Institution: Institut Teknologi Bandung

Physics is the science that studies the properties and behavior of matter and energy in the universe. This science explains various natural phenomena through observation, experiments, and mathematical analysis. Physics combines theoretical and practical concepts to understand how the universe operates. Econophysics is a branch of physics that integrates physics concepts and methods with economics to study economic phenomena. This research aims to compare stock price prediction outcomes for August 2023 using Monte Carlo simulations with Long-Short Term Memory machine learning models. The research begins with a literature review related to Monte Carlo simulation and Long-Short Term Memory. Subsequently, a Monte Carlo simulation model is created using the concept of Geometric Brownian Motion, as well as Long-Short Term Memory, in the Python programming language. The Monte Carlo simulation model that is constructed will be run through various iterations to observe the distribution of stock price predictions for Bank BCA, Unilever, and Telkom Indonesia. The Long-Short Term Memory model will be executed with the same parameters for each of the Bank BCA, Unilever, and Telkom Indonesia stocks. The results of the Monte Carlo simulation and Long-Short Term Memory modeling for Bank BCA, Unilever, and Telkom Indonesia consist of graphs depicting the predicted stock prices compared to the actual stock price graphs and the Root Mean Square Error (RMSE) values. The RMSE values obtained for Bank BCA is in the range 72 – 195, for Unilever is in the range 37 – 133, and for Telkom Indonesia is in the range 40 – 164.