Quantile regression estimation for network GARCH(1,1) model

The generalized autoregressive conditional heteroscedasticity (GARCH) model is common in time series. The network GARCH(1,1) model is constructed by introducing network structure into the traditional GARCH(1,1) model, which significantly improves the performance of modeling. In this paper, we consider the quantile regression estimation for parameters in the network GARCH(1,1) model, establish the asymptotic properties of the parameter estimators, and give the numerical analysis for related estimators. The closing prices of 21 banks are constructed to verify the practicability of the considered model.