چكيده به لاتين
Time series is one of the statistical and probability indicators that has many applications in various fields such as Geophysics , Communication engineering , Financial engineering , Meteorology , Economics , Medical sciences , Biology , Psychology , Astronomy , Social sciences and other sciences.
The introduction of time series statistics started from the time series of one variable and eventually led to vector time series. As the stock market is effective in country's finance market, finding a suitable way to predict the stock market is very important. For this reason, the aim is to evaluate the predictive power of linear models in the stock market. Of course, data are analyzed using financial time series analysis (fluctuating data analysis and market microstructure.) Fluctuating data include observations taken at very short intervals. Fluctuating data, in financial sciences, are trade-to-trade data in stock markets .Here time is measured in seconds. Fluctuating financial data are important in the study of transverse issues related to the business process and market microstructure. High- volatility data has unique properties that do not appear in low-volatility data.
Due to the possibility of nonlinear relationship in financial markets, the purpose of this paper is to evaluate the predictive power of linear and nonlinear models in the stock market. In this dissertation we focus on the nonlinear models in the analysis of financial time series. We introduced some nonlinear models that are applicable in financial time series.
