Stock price modeling and forecasting using meta-heuristic ant colony algorithm
The ability to predict prices in the capital markets has always had supporters and opponents in wide ranges. But the empirical evidence shows that the price in the financial markets is somewhat predictable, but achieving a proper forecast requires knowledge of non-linear patterns and the ability to predict the market's memory. The current research is an applied research, the purpose of which is to model and predict stock price predictions in the capital market using non-linear algorithms. To achieve this goal, the total stock price index data has been used in the period of 2016 to 2021 and on a monthly basis. The data have been reviewed after collection using the smoothing method for holidays, and in order to increase the accuracy of the models, the optimal window length of each algorithm has been calculated. The findings show that the ant algorithm has a very good ability to model and predict the price in the capital market by minimizing the prediction error. Also, this algorithm is faster in achieving the optimal price in a six-month period compared to the genetic algorithm.
-
The Relationship between Board Gender Diversity and Real and Accrued Earnings Management Practices
, Mehdi Mehrabi Zadeh *, Zahra Mohammadi
Journal of New research approaches in management and accounting, -
Comparative application of particle algorithm and genetic algorithm in predicting long-term and short-term trend of stock returns
Javad Kiae, *
Journal of Financial Economics,