The double-threshold autoregressive conditional heteroscedastic(DTARCH) model is a useful tool to measure and forecast the mean and volatility of an asset return in a financial time series. The DTARCH model can handle...The double-threshold autoregressive conditional heteroscedastic(DTARCH) model is a useful tool to measure and forecast the mean and volatility of an asset return in a financial time series. The DTARCH model can handle situations wherein the conditional mean and conditional variance specifications are piecewise linear based on previous information. In practical applications, it is important to check whether the model has a double threshold for the conditional mean and conditional heteroscedastic variance. In this study, we develop a likelihood ratio test based on the estimated residual error for the hypothesis testing of DTARCH models. We first investigate DTARCH models with restrictions on parameters and propose the unrestricted and restricted weighted composite quantile regression(WCQR) estimation for the model parameters. These estimators can be used to construct the likelihood ratio-type test statistic. We establish the asymptotic results of the WCQR estimators and asymptotic distribution of the proposed test statistics. The finite sample performance of the proposed WCQR estimation and the test statistic is shown to be acceptable and promising using simulation studies. We use two real datasets derived from the Shanghai and Shenzhen Composite Indexes to illustrate the methodology.展开更多
To test variance homogeneity,various likelihood-ratio based tests such as the Bartlett's test have been proposed.The null distributions of these tests were generally derived asymptotically or approximately.We re-e...To test variance homogeneity,various likelihood-ratio based tests such as the Bartlett's test have been proposed.The null distributions of these tests were generally derived asymptotically or approximately.We re-examine the restrictive maximum likelihood ratio(RELR)statistic,and sug-gest a Monte Carlo algorithm to compute its exact null distribution,and so its p-value.It is much easier to implement than most existing methods.Simulation studies indicate that the proposed procedure is also superior to its competitors in terms of type I error and powers.We analyse an environmental dataset for an illustration.展开更多
基金supported by National Natural Science Foundation of China(Grant No.71601123)MOE(Ministry of Education in China)Project of Humanities and Social Sciences(Grant No.15YJC910004)+3 种基金supported by National Natural Science Foundation of China(Grant No.11471277)the Research Grant Council of the Hong Kong Special Administration Region(Grant No.GRF14305014)supported by the State Key Program of National Natural Science Foundation of China(Grant No.71331006)the Major Research Plan of National Natural Science Foundation of China(Grant No.91546202)
文摘The double-threshold autoregressive conditional heteroscedastic(DTARCH) model is a useful tool to measure and forecast the mean and volatility of an asset return in a financial time series. The DTARCH model can handle situations wherein the conditional mean and conditional variance specifications are piecewise linear based on previous information. In practical applications, it is important to check whether the model has a double threshold for the conditional mean and conditional heteroscedastic variance. In this study, we develop a likelihood ratio test based on the estimated residual error for the hypothesis testing of DTARCH models. We first investigate DTARCH models with restrictions on parameters and propose the unrestricted and restricted weighted composite quantile regression(WCQR) estimation for the model parameters. These estimators can be used to construct the likelihood ratio-type test statistic. We establish the asymptotic results of the WCQR estimators and asymptotic distribution of the proposed test statistics. The finite sample performance of the proposed WCQR estimation and the test statistic is shown to be acceptable and promising using simulation studies. We use two real datasets derived from the Shanghai and Shenzhen Composite Indexes to illustrate the methodology.
基金The research of Li was supported by Grant 11871294 from National Natural Science Foundation of ChinaLiang’s research was partially supported by NSF grant DMS-1620898.
文摘To test variance homogeneity,various likelihood-ratio based tests such as the Bartlett's test have been proposed.The null distributions of these tests were generally derived asymptotically or approximately.We re-examine the restrictive maximum likelihood ratio(RELR)statistic,and sug-gest a Monte Carlo algorithm to compute its exact null distribution,and so its p-value.It is much easier to implement than most existing methods.Simulation studies indicate that the proposed procedure is also superior to its competitors in terms of type I error and powers.We analyse an environmental dataset for an illustration.
