One of the most popular methods of being involved in the capital market is investment in the stocks exchange via offered shares, as one of possibly profitable options. Stock market has a nonlinear and chaotic behavior which is affected by psychological, economic, and political conditions. To predict share price, nonlinear intelligent methods such as learning automata, metaheuristics (firefly algorithm), and artificial neural network can be used. As part of our research project, a model for predicting the share price has developed and presented by combining these methods. Information of 20 companies listed on Tehran Stock Exchange has been predicted and presented in order to evaluate the proposed solution. This algorithm can approximately predict 30 to 50% of the future behavior of shares in the market. The outcomes have shown great potential for the hybrid algorithm to be applied in other fields of business administration.
Autonomic Learning Algorithm to Predict Stock Price at the Exchange via Metaheuristics-Based Optimization
8 Views
6 Downloads