Non-Linear Non-Stationary Heteroscedasticity Volatility for Tracking of Jump Processes
share
In this paper, we introduce a new jump process modeling which involves a particular kind of non-Gaussian stochastic processes with random jumps at random time points. The main goal of this study is to provide an accurate tracking technique based on non-linear non-stationary heteroscedasticity (NNH) time series. It is, in fact, difficult to track jump processes regarding the fact that non-Gaussianity is an inherent feature in these processes. The proposed NNH model is conditionally Gaussian whose conditional variance is time-varying. Therefore, we use Kalman filter for state tracking. We show analytically that the proposed NNH model is superior to the traditional methods. Furthermore, to validate the findings, simulations are performed. Finally, the comparison between the proposed method and other alternatives techniques has been made.
READ FULL TEXT