• O. E. Taylor
• P. S. Ezekiel
• C. G. Igiri

The financial market has become an essential market in today’s economy. It is a market where different commodities are being exchange or sold at a certain price, though prices of commodities in the stock market are not stable. They are being influenced by some external factors like, politics, natural calamities, investor’s sentiment, exchange rate e.t.c.This system propose an agent using Reinforcement Learning Technique to predict google stock market data. The agent was trained using three Reinforcement Learning Algorithms. The algorithms used in this system are Deep Q-Learning Network, Double Q-Learning Network and Dueling Double Q-Learning Network. We first create an environment of which the agent can learn from. The environment consist of state, action and reward. The state is the representation of the environment; the action is the movement of the agent within the state or the choice/decision made by the agent at a given state. The action taken by the agent are stay, buy or sell. Our experimental results shows that Double Q-Learning Network outperforms the Deep Q-Learning Network and Dueling Double Q-Learning Network by receiving the highest reward in each state that the agent tries to take the best decision that is to say, buying the movement of the market is an upward trend, selling when the movement of the market is a downward trend and holding back when the movement of the market is unstable. The Dueling Q-Learning Network had the best reward of about 29.4 on 35 epoch, Double Q-Learning had reward of about 35.4 on 45 epoch, and the Deep Q-Learning Network had reward of about -81.8 on the first epoch.

Reinforcement Learning, Deep Q-learning Network, Double Q-Learning Network, Dueling Q-Learning, Google Stock Market data

"Google Stock Market Price Prediction using Reinforcement Learning Technique", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 1, page no.71-81, January-2021, Available :http://www.jetir.org/papers/JETIR2101213.pdf

"Google Stock Market Price Prediction using Reinforcement Learning Technique", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 1, page no. pp71-81, January-2021, Available at : http://www.jetir.org/papers/JETIR2101213.pdf