Application Of Composite Quantile Regression Method In Time Series Model Of National Debt Risk

Based on the importance of time series analysis and the many advantages of quantile regression estimation,the quantile regression series estimation method is applied to the time series model to explore more effective estimation methods for time series models,and the consistency of estimation methods is studied.And discuss and apply the model to the quantification of risk in the national debt market.Different from other financial products,China's national debt market is affected by both political and economic factors.For risk fluctuations with special significance,proper risk prediction is particularly important.In this paper,we use composite(CQR)and weighted composite(WCQR)quantile regression as the research object,and explore the effective estimation of ARIMA time series model,and realize the reasonable prediction of the sudden risk situation of China's national debt market.The main contents include:1.The quantile regression(quantitative(QR),composite(CQR)and weighted composite(WCQR))estimators of ARIMA model are established;the asymptotic normality of ARIMA model under different quantile regression is discussed.The weighted composite quantile regression optimal weight estimation is obtained.2.By analyzing the large fluctuations in the national debt market in China,from the perspective of quantile time series regression,we obtain a data that can reflect the fluctuation range in the fluctuation period,and estimate the fluctuation value of this data by using the composite quantile regression estimation method.Through the obtained values,the quantitative analysis of the large fluctuations in the national debt market during the Sino-US trade war was conducted.3.For the data during the Sino-US trade war,using the weighted composite quantile regression estimation method,according to the length of the study time,select the same bin intercept or different bin intercepts for modeling analysis,and give Sino-US trade.Effective analysis and accurate prediction of risk impacts before and after the war.