Study On Application Of Nonlinear Tracking-Differentiator In VaR

The core and foundation of risk management is risk measurement. Because of its comprehensive measurement, VaR system becomes a major method for financial market risk measurement. Although the concept of VaR is simple, its measurement is a statistical problem with challenge. Western and eastern scholars went deep into discussions around the calculation of VaR. Usually it has three kinds of major computing technologies, and every kind of VaR computing technology has its weak points. People have put forward a lot of improved methods to this in recent years."Economic physics" that rose gradually in the middle of the 1990s has provided a new visual angle for major economics research. In view of this visual angle, the complicated science research approach begins to enter the fields of economics and finance. This new way of Natural scientific method spreading to social economic field offers a new train of thought for the research of this text. The author will utilize Nonlinear Tracking-Differentiator (abbreviated as NTD) in cybernetics to simulate and predict financial time series, and attempt to put forward a kind of new emulation simulation method, applying to the calculation of VaR.The main research work of this text is as follows:This text introduces VaR model frame at first, discusses in details the three basic computing technologies of VaR, and then compares them and finds their weak points and places that can be improved. In what follows it presents the general theory of NTD and several kinds of concrete NTDs, derives out the Dispersed form of the two order fast-controlling system NTD and carries on numerical emulation, then applies it to VaR calculation and gets Nonlinear Tracking-Differentiator simulation VaR. Finally, it carries on the empirical research of the Nonlinear Tracking-Differentiator Simulation VaR Risk Measurement Technology in China's Securities Market. To the three concrete VaR computing technologies (Monte Carlo Simulation VaR, Historical Simulation VaR, Nonlinear Tracking-Differentiator simulation VaR), it also carries on comparative analysis and appraise.