Proposition of a model For Forecasting Value at Risk in One Step Ahead

Subject Areas : Financial engineering

ehsan Mohammadian Amiri ¹ , S. Babak Ebrahimi ² , maryam Nezhad Afrasiabi ³

1 - MSc. Student in Financial Engineering, Faculty of Industrial Engineering, K.N.Toosi University of Technology, Tehran, Iran
2 - Assistant Prof., Faculty of Industrial Engineering, K.N.Toosi University of Technology, Tehran, Iran
3 - MSc. Student in Industrial Engineering, Faculty of Industrial Engineering, K.N.Toosi University of Technology, Tehran, Iran

Received: 2017-04-20 Accepted : 2017-06-09 Published : 2016-09-22

Keywords: Value at risk, forecasting future periods, the exponential smoothing family, Justified exponential smoothing,

Abstract :

Risk forecasting for future periods plays an important role in making the right decisions of managers and financial activists to invest in companies and institutions. On the other hand wrong decisions of commercial managers can have undesirable consequences for their organizations. Therefore the most important issues for investors is forecasting risk in future periods. The importance of this issue was caused us to forecast Value at Risk (VaR) in one step ahead by using the exponential smoothing family for two normal and t-student distributions with confidence levels of 95%, 97.5% and also 99% in this research. Previously the classic method is commonly used to forecast future periods of VaR, but in this research the family of exponential smoothing models is used, which process data by considering trend and doing so online monitoring. In order to validate the model, the proposed model has been compared with the classic method by using backtesting. The results confirms the more accurate forecasting of proposed method in normal distribution with confidence levels of 97.5%, and 99% and also in t-student distribution with confidence levels of 97.5%, 99%.