Majumdar, Mahbubul AlamDutta, AmitParvez, Md SultanTalukdar, Partho2025-09-302025-09-3020202020-10ID 16101100ID 16101079ID 16101095http://hdl.handle.net/10361/26808Cataloged from PDF version of thesis.Includes bibliographical references (pages 43-44).This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science, 2020.The purpose of this work is to create an agent that can trade efficiently in the stock market. There is an implementation,proximal policy optimization (PPO) to train the agent and OpenAIGym to simulate a finical Market environment. The biggest problem of the trading market is there is no specific trading strategies, more often investor focuses on the risk and thus it becomes more of gambling. The deep learning community find research on Financial market less interesting because of the difficulty and the expensive nature of financial market. Main goal is to introduce a trading model using Reinforcement learning and neural network. The model will create a better solution of the current anomaly. The process gives confidence that this model will help the investor to find a safe yet profitable strategy.50 pagesenBRAC University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Reinforcement learningMachine learningProximal policy optimizationTrading indicatorsOpenAIGymTrading marketFinancial marketNeural networksRNNDNNLSTMReinforcement learning.Finance--Mathematical models--Data processing.Neural networks (Computer science).Thesis