16483

14666

كارشناسي ارشد

آمار اقتصادي و اجتماعي

اصفهان : دانشگاه صنعتي اصفهان

1399

هشت، 107ص.: مصور، جدول، نمودار

صفيه محمودي

مهدي خاشعي

واژ ه نامه

افشين پرورده، ريحانه ريخته گران

1400/02/20

كتابنامه

آمار

رياضي

1400/03/11

2696170

108 فهرست نمادها On the Properties of GARCH Class Models and Their Stock Market Application ZAHRA ASGARIAN z asgariyan@math iut ac ir February 8 2021 Master of Science Thesis in Farsi Departement of Mathematical Sciences Isfahan University of Technology Isfahan 84156 8311 IranSupervisor Dr Safieh Mahmoodi mahmoodi@iut ac irAdvisor Dr Mehdi Khashei khashei@iut ac ir2000 MSC 10M62 20P62 84B91Keywords Time Series Return Volatility Cluster Volatility Heteroskedasticity Conditional Variance Garch Models Abstract This M Sc thesis is based on the following papers GARCH 101 The use of ARCH GARCH models in applied econometrics Journal of E R economic perspectives 15 no 4 2001 157 168 I Volatility Modelling using Arch and Garch Models A Case Study of the A K C U P Nigerian Stock Exchange International Journal of Mathematics Trends and Technology IJMTT Volume 65 Issue 4 April 2019Forecasting financial markets has many applications in risk management and monetary policy decisions and selecting the best type of stock has always been an issue for the investors For forecasting the corre sponding model should be obtained and the statistical model should contain the important information aboutthe ROI Return on Investment The widely used classic models such as ARMA models have centralizedstructures and since the second order structure in most financial time series data is close to the second or der of the white noise and do not provide any forecasting information it is not appropriate to use them In addition the Moving Average MA and the autoregressive AR and the ARMA have homogeneityof variance while investigating the financial data it was concluded that the homogeneity of variance isviolated There are great fluctuations between the periods in economic and financial variables such as theoil price The risks are greater is some periods That is the predicted error in some periods is greater Nonetheless the high risk volatile times are not randomly scattered Accordingly a model is requiredwith no need to the homogeneity of variances ARCH and GARCH models are designed for such se ries Instead of using the standard deviation of the great and small samples these models suggest that theweighted mean of the predicted errors be considered as a weighted variance these weights can have greatereffects on the recent information and less effects on the previous information Therefore these models arewidely used tools for series with heterogeneity The purpose of using these models is to obtain a methodfor measuring the fluctuations such as the standard deviation that will be used in financial decisions madebased on risk analysis stock selection and pricing This class includes several models any of which could be used in regard to data features For instance EGARCH model is a non linear model which is more suited to data with leverage effect In order to examine and compare models information criteria such as Akaike and Schwarz or forecastcriteria will be used It should be note some of the that most references consider Schwarz criterion moreproper than other information criteria

صفيه محمودي

مهدي خاشعي

افشين پرورده، ريحانه ريخته گران