Investigation of Fractal Property Price and Stock Returns of Tehran Stock Exchange Companies Using Nonlinear ARIFMA Model

Much evidence suggests that time series such as stock market prices are complex and random, which makes their changes unpredictable. However, these time series are likely to be a nonlinear dynamic or, in other words, a chaotic process and can therefore be predictable. Therefore, in this study, stock prices and stock returns of Tehran Stock Exchange companies during the period 2014-2018 and monthly intervals were tested to determine whether these variables have fractal properties in their behavior. To achieve the above objective, our model estimation is used to explain the mass fraction of moving average. The findings of the above tests indicate that stock prices and stock returns experience a turbulent and definite process. This implies that the capital market is inefficient, and because of its long-term memory, it can be useful in predicting long-term performance and may have a guide to better understanding market failure factors such as the lack of transparency of information flow and action to address it.