SOLUTION OF SEMIPARAMETRIC MODEL WITH DERAILIZED LEAST SQUARE AND STATISTIC CHARACTERISTICS OF ESTIMATOR
Ding Shijun 1,3) ; and Jiang Weiping 2)
1)School of Geodesy and Geomation, Wuhan University, Wuhan 4300792)GNSS Engineering Research Center of Wuhan University, Wuhan 4300793)Key Laboratory of Precise Engineering and Industry Surveying State Bureau of Surveying and Mapping,\= Wuhan 430079
Abstract：The semiparametric model is a combination of classical linear model and nonparametric regression model. The two didfferent computing methods on the resolution of the semiparametric model under the principle of penalized least squares are discussed and the computation formulas of the resolutions and estimate accuracy are detailed.On the basis of the statistic characteristics of random error，the statistic characteristics of the semiparametric model estimators are discussed about mathematical expection,variances of estimates and mean square error of the parameter estimate.Finally, a simulated example is demonstrated and some conclusions are drawn.
Ding Shijun,and Jiang Weiping . SOLUTION OF SEMIPARAMETRIC MODEL WITH DERAILIZED LEAST SQUARE AND STATISTIC CHARACTERISTICS OF ESTIMATOR[J]. jgg, 2010, 30(6): 47-50.
