Trading Agent for the Indian Stock Market scenario using Actor-Critic based Reinforcement Learning

Financial trading is about buying, holding, and selling securities in the hope of making a profit. Automation is a trending area in the engineering domain that can help maximize the outcome of interest. Machine learning approaches like reinforcement learning have a great potential to solve automation in certain business domains, thereby maximizing the work outcome. Reinforcement learning has the sole objective of attaining maximum profit or reward in the given environment where the agent acts. Hence, the proposed work deals with leveraging the state of the art Actor-Critic Reinforcement learning algorithms like Proximal Policy optimization (PPO), Deep Deterministic Policy Gradient (DDPG), Advantage Actor Critic (A2C), and Twin Delayed DDPG (TD3) in developing a trading agent which can make decisions based on the trading environment whether the stock has to be bought, sold or held at the given instant in the Indian Stock Market scenario.