Predicting the Overall Index of Tehran Stock Exchange UsingSingular spectrum analysis Based on Genetic Algorithm and Overlap Singular spectrum analysis
Predicting the Overall Index of Tehran Stock Exchange Using Singular spectrum analysis Based on Genetic Algorithm and Overlap Singular spectrum analysis
Subject Areas : Financial Knowledge of Securities Analysis
zahra حسن دوست 1 , hamidreza vakilifard 2 , فریدون رهنمای رودپشتی 3
1 - Department of financial managemet,management and economics,science and research branch,islamic azad university, tehran,iran
2 - Associate Professor of Financial Management, Islamic Azad University, Tehran, Iran.
3 - نویسنده مسئول
Keywords: ", Window Length", , ", Signal", , ", Break Point", , ", Noise", , ", Overlap Singular spectrum analysis", ,
Abstract :
The present study analyzes the prediction of the total index of the Tehran Stock Exchange with the singular spectral analysis method based on the genetic algorithm and overlapping singular spectral analysis. It is practical in terms of purpose and descriptive-analytical in terms of method. Its statistical population is the daily price of the total index of the Tehran Stock Exchange in the ten-year return (2009 to 2018) and the research sample is 2411 data from the logarithmic return of the target index.First, the genetic algorithm was implemented in the SSA method on the index. Then, using the Ov-SSA method in order to improve the reconstruction and resolution of the components, the initial and large time series were divided into small and common consecutive parts and the standard SSA analysis method was used for each part.The results of the research showed that the Ov-SSA analysis method has a higher performance than the GA-SSA analysis method with a lower mean absolute value error.