Online learning is a general framework for sequential decision-making under uncertainty. On each round, the learner performs an action and observes some feedback from the environment.
We apply the Online Learning framework to the context of stock exchange, what is called online portfolio selection. In this setting, at each round the learner chooses an allocation vector, named portfolio, which specifies how the money should be invested. Hence, the feedback is the return of investment at the end of the trading session, e.g., a minute or a day.
In this project, we are interested in the case of a single stock. Therefore, we have two investment opportunities – long or short the stock. This prediction problem turns out to be quite challenging when applied in reality with true stock prices and their high volatility.
Based on this premise, we construct a new framework, similar to the online portfolio selection. In this framework, the portfolio is selected according to upper and lower thresholds on the stock price. The thresholds should evaluate if the stock is priced high, low or correctly on the market.
Finally, we tackle the problem using Prediction with Expert Advice. In our case, the expert information will be the correct thresholds for the stock price. This way, we expect to minimize the regret. We show the exploration-exploitation trade-off applied to a random process. Lastly, we propose future research ideas.