Research On The Risk Measurement Of Interbank Offerd Rate Based On EGARCH-POT Model

Since the central bank in June 1996 release interbank interest rates, to 2015 October the central bank announced the release of the ceiling on deposit interest rates control, after twenty years, with a series of large and small banks relax control policy implementation,the Chinese interest rate marketization reform basically completed. With the continuous reform of interest rate market in China, the market interest rate fluctuations become more and more frequent, the interest rate risk is increasing. Financial assets are the main assets of commercial banks, the frequent fluctuations in interest rates will lead to changes in the value of their assets, so that the interest rate risk of commercial banks increased. Since China has implemented a long period of interest rate control policy, commercial banks mainly credit risk management in the first place. With the constant deepening of the interest rate market reform, the interest rate risk has continued to highlight, to strengthen the management of interest rate risk of commercial banks is of great significance.On the interest rate risk of commercial banks of the metric, due to historical and technical reasons, China’s commercial banks mostly choose Static interest rate risk measurement methods such as interest rate sensitive gap analysis, duration and convexity analysis; The market risk management model based on VaR is the future development direction of China’s commercial banking market risk capital measurement. The calculation method of VaR model can be divided into three kinds, namely the parameter method, the non parametric method and the semi parametric method. The parameter method needs to simulate the probability distribution of the return on assets, assuming that the assets return rate obeys a certain distribution, so it is easy to have the setting error of the model; Non parametric method is based on a large amount of historical data without the need for simulation yields the probability distribution can better avoid the risk model specification errors, but easily lead to "dimension disaster" problem and model results is difficult to explain, mainly is the historical simulation method and Monte Carlo simulation method.The semi parametric method is combined with the non parametric method to overcome the error of the model and the "Curse of dimensionality". The main methodsinclude the extreme value theory, the first four moments method and the semi parametric-GARCH method.This paper is mainly based on the parametric and semi parametric method to establish model to solve VaR value. Firstly, based on the traditional theory of establishing GARCH model, residuals are assumed to obey three different distributions of normal distribution,distribution, GED distribution, establish GARCH model, EGARCH model and TGARCH model a total of 9 model; Because the traditional VaR is selecting a high quantile and easy to overlook the thick tail distribution caused by extreme events. This paper introduces the extreme value theory POT model to solve the problem of tail risk, combined with EGARCH model and extreme value POT model to build EGARCH-POT model, obtaining the EGARCH model for residual sequence of a pot model of extreme value to calculate the VaR. This paper selects 2009 January to August 2015 market degree higher bank of Shanghai interbank offered rate(SHIBOR) data for analysis and research, selecting the failure rate test and kupiec back testing method of the model calculation results were compared and analyzed.The empirical results show that: under 95% and two 99% different confidence levels, the EGARCH-POT model can better fit the distribution of yield, and it is more suitable to measure the risk of China’s commercial banks offed rate.