In this paper the limiting distribution of the least square estimate for the autoregressive coefficient of a nearly unit root model with GARCH errors is derived. Since the limiting distribution depends on the unknown ...In this paper the limiting distribution of the least square estimate for the autoregressive coefficient of a nearly unit root model with GARCH errors is derived. Since the limiting distribution depends on the unknown variance of the errors, an empirical likelihood ratio statistic is proposed from which confidence intervals can be constructed for the nearly unit root model without knowing the variance. To gain an intuitive sense for the empirical likelihood ratio, a small simulation for the asymptotic distribution is given.展开更多
Background:Modeling exchange rate volatility has remained crucially important because of its diverse implications.This study aimed to address the issue of error distribution assumption in modeling and forecasting exch...Background:Modeling exchange rate volatility has remained crucially important because of its diverse implications.This study aimed to address the issue of error distribution assumption in modeling and forecasting exchange rate volatility between the Bangladeshi taka(BDT)and the US dollar($).Methods:Using daily exchange rates for 7 years(January 1,2008,to April 30,2015),this study attempted to model dynamics following generalized autoregressive conditional heteroscedastic(GARCH),asymmetric power ARCH(APARCH),exponential generalized autoregressive conditional heteroscedstic(EGARCH),threshold generalized autoregressive conditional heteroscedstic(TGARCH),and integrated generalized autoregressive conditional heteroscedstic(IGARCH)processes under both normal and Student’s t-distribution assumptions for errors.Results and Conclusions:It was found that,in contrast with the normal distribution,the application of Student’s t-distribution for errors helped the models satisfy the diagnostic tests and show improved forecasting accuracy.With such error distribution for out-of-sample volatility forecasting,AR(2)–GARCH(1,1)is considered the best.展开更多
Temperature （T） and salinity （S） profiles from conductivity-temperature-depth data collected during the Northern South China Sea Open Cruise from August 16 to September 13, 2008 are assimilated using Ensemble Kalm...Temperature （T） and salinity （S） profiles from conductivity-temperature-depth data collected during the Northern South China Sea Open Cruise from August 16 to September 13, 2008 are assimilated using Ensemble Kalman Filter （EnKF）. An adaptive observational error strategy is used to prevent filter from diverging. In the meantime, aiming at the limited improvement in some sites caused by the T and S biases in the model, a T-S constraint scheme is adopted to improve the assimilation performance, where T and S are separately updated at these locations. Validation is performed by comparing assimilated outputs with independent in situ data （satellite remote sensing sea level anomaly （SLA）, the OSCAR velocity product and shipboard ADCP）. The results show that the new EnKF assimilation scheme can significantly reduce the root mean square error （RMSE） of oceanic T and S compared with the control run and traditional EnKF. The system can also improve the simulation of circulations and SLA.展开更多
This paper implements the method of estimating functions (EF) in the modelling and forecasting of financial returns volatility. This estimation approach incorporates higher order moments which are common in most finan...This paper implements the method of estimating functions (EF) in the modelling and forecasting of financial returns volatility. This estimation approach incorporates higher order moments which are common in most financial time series, into modelling, leading to a substantial gain of information and overall efficiency benefits. The two models considered in this paper provide a better in-sample-fit under the estimating functions approach relative to the traditional maximum likely-hood estimation (MLE) approach when fitted to empirical time series. On this ground, the EF approach is employed in the first order EGARCH and GJR-GARCH models to forecast the volatility of two market indices from the USA and Japanese stock markets. The loss functions, mean square error (MSE) and mean absolute error (MAE), have been utilized in evaluating the predictive ability of the EGARCH vis-à-vis the GJR-GARCH model.展开更多
基金Supported by the National Natural Science Foundation of China(10801118)Specialized Research Fund for the Doctor Program of Higher Education(200803351094)
文摘In this paper the limiting distribution of the least square estimate for the autoregressive coefficient of a nearly unit root model with GARCH errors is derived. Since the limiting distribution depends on the unknown variance of the errors, an empirical likelihood ratio statistic is proposed from which confidence intervals can be constructed for the nearly unit root model without knowing the variance. To gain an intuitive sense for the empirical likelihood ratio, a small simulation for the asymptotic distribution is given.
文摘Background:Modeling exchange rate volatility has remained crucially important because of its diverse implications.This study aimed to address the issue of error distribution assumption in modeling and forecasting exchange rate volatility between the Bangladeshi taka(BDT)and the US dollar($).Methods:Using daily exchange rates for 7 years(January 1,2008,to April 30,2015),this study attempted to model dynamics following generalized autoregressive conditional heteroscedastic(GARCH),asymmetric power ARCH(APARCH),exponential generalized autoregressive conditional heteroscedstic(EGARCH),threshold generalized autoregressive conditional heteroscedstic(TGARCH),and integrated generalized autoregressive conditional heteroscedstic(IGARCH)processes under both normal and Student’s t-distribution assumptions for errors.Results and Conclusions:It was found that,in contrast with the normal distribution,the application of Student’s t-distribution for errors helped the models satisfy the diagnostic tests and show improved forecasting accuracy.With such error distribution for out-of-sample volatility forecasting,AR(2)–GARCH(1,1)is considered the best.
基金The Strategic Priority Research Program of the Chinese Academy of Sciences under contract No.XDA10010405the Promgram of Guangdong Province Department of Science and Technology No.2012A032100004+1 种基金the National Natural Science Foundation of China under contract Nos 41476012,41521005 and 41406131the Knowledge Innovation Program of the Chinese Academy of Sciences under contract Nos SQ201001 and SQ201205
文摘Temperature （T） and salinity （S） profiles from conductivity-temperature-depth data collected during the Northern South China Sea Open Cruise from August 16 to September 13, 2008 are assimilated using Ensemble Kalman Filter （EnKF）. An adaptive observational error strategy is used to prevent filter from diverging. In the meantime, aiming at the limited improvement in some sites caused by the T and S biases in the model, a T-S constraint scheme is adopted to improve the assimilation performance, where T and S are separately updated at these locations. Validation is performed by comparing assimilated outputs with independent in situ data （satellite remote sensing sea level anomaly （SLA）, the OSCAR velocity product and shipboard ADCP）. The results show that the new EnKF assimilation scheme can significantly reduce the root mean square error （RMSE） of oceanic T and S compared with the control run and traditional EnKF. The system can also improve the simulation of circulations and SLA.
文摘This paper implements the method of estimating functions (EF) in the modelling and forecasting of financial returns volatility. This estimation approach incorporates higher order moments which are common in most financial time series, into modelling, leading to a substantial gain of information and overall efficiency benefits. The two models considered in this paper provide a better in-sample-fit under the estimating functions approach relative to the traditional maximum likely-hood estimation (MLE) approach when fitted to empirical time series. On this ground, the EF approach is employed in the first order EGARCH and GJR-GARCH models to forecast the volatility of two market indices from the USA and Japanese stock markets. The loss functions, mean square error (MSE) and mean absolute error (MAE), have been utilized in evaluating the predictive ability of the EGARCH vis-à-vis the GJR-GARCH model.
