AGENT-BASED CELLULAR AUTOMATA SIMULATION FOR INVESTOR BEHAVIOR ANALYSIS IN STOCK MARKET

• Main Author: Safly Ramdhani, Nabila
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/65153
• Institution: Institut Teknologi Bandung

Physics is a field of natural science that studies the universe and its materials, include study the behavior of living things like humans who carry out certain activities in various aspects of life, including social, cultural, and economic activities. The existence of an economy that is growing in modern times makes people increasingly aware of the importance of managing finances by investing, including stock investment. Stock investment are very popular because they require only a small amount of capital but with potentially big profits. In order to gain profits in stock investing, investors have their own strategies in determining their transaction decisions. This investor behavior can be modeled to see and predict the stock prices that forms in the market as a way to determine investment strategies. In this study, a simulation using agent-based modeling based on cellular automata was made to analyze the behavior of investors in the stock market from the stock price movements. The stock prices that are formed from two types of investors, namely followers and contrarians are analyzed to see the effect of investor behavior in stock market. In addition to the stock price movements, the stock returns and volatility can also be seen from all variations of the simulation. An interesting phenomenon is found from the simulation where the market in a stable condition with the value of volatility below 0.5% occurs when the ratio of the number of investors is balanced in the market. An optimal stock market have not yet been achieved, but the simulations can create suitable rules that are similar to the stock market in the real world.