Using the Wavelet neural networks to determine and evaluate the effects of systematic risk on financial returns of stock

This paper examines the relationship between stock returns and the systematic risk in the medium and long term time horizons and evaluates the effect amount of market fluctuations on the above relationship in nutrients and dairy companies listed in the Tehran Stock Exchange during the 2008 to 2013. In the financial market capital is one of the most important issues of the relationship between risk and returns, especially systematic risk; Because it is believed that stock returns is only a function of the systemic risk. Several studies have been conducted to investigate the relationship between risk and returns. Among these efforts, is a study that conducted by Sharp. He by introducing the model of (CAPM Capital Asset Pricing Model) assumed that there is a simple linear and positive relationship between systematic risk and exchange return. In order to testing the research hypothesis at first the research period based on the variance for the pharmaceutical and chemical industry index has been divided into two categories of high oscillation and low oscillation periods. Then information related to systematic risk and stock returns in the periods of high oscillation and low oscillation overlap by method of Discrete wavelet transform (DWT) and by Daubechies  wavelet by using MATLAB software parse to smaller period of time, then regression analysis was used  in order to testing  research hypotheses. Results of testing hypotheses indicate  there is a significant relationship between systematic risk and returns in the high oscillation time period in the medium and long term time horizons. Also in time periods of low oscillation also in the medium-term time horizons there is a significant relationship between systematic risk and returns but meaningful relationship between risk and return is approved only in long-term time horizon for 182 days.