For the linear model y_i=x_iθ+e_i, i=1, 2,…, let the error sequence {e_i}_i=1 be iidr.v.’s, with unknown density f(x). In this paper,a nonparametric estimation method based onthe residuals is proposed for estimatin...For the linear model y_i=x_iθ+e_i, i=1, 2,…, let the error sequence {e_i}_i=1 be iidr.v.’s, with unknown density f(x). In this paper,a nonparametric estimation method based onthe residuals is proposed for estimating f(x) and the consistency of the estimators is obtained.展开更多
The profile error evaluation of complex curves and surfaces expressed inparametric form is considered. The linear error model is established on the base of two hypothesesfirstly. Then the profile error evaluation is c...The profile error evaluation of complex curves and surfaces expressed inparametric form is considered. The linear error model is established on the base of two hypothesesfirstly. Then the profile error evaluation is converted into one of these optimal formulations:MINIMAX, MAXMIN and MINIDEX problems, which are easier to be solved than the initial form. To eachone of them, geometric condition and algebraic condition are presented to arbitrate whether theideal element reaches to the optimal position. Exchange algorithm is proven highly effective insearching for solutions to these optimization problems. At last some key problems in tolerance offreeform surfaces and curves in B spline method are discussed.展开更多
This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed...This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data.Under the structural density assumption,the C-step technique borrowed from the Rousseeuw＇s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization.To eliminate the model ambiguities of the multiple-structural data,statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation.Experiments show that the efficiency and robustness of the proposed algorithm.展开更多
基金The project supported by National Natural Science Foundation of China Crant 18971061
文摘For the linear model y_i=x_iθ+e_i, i=1, 2,…, let the error sequence {e_i}_i=1 be iidr.v.’s, with unknown density f(x). In this paper,a nonparametric estimation method based onthe residuals is proposed for estimating f(x) and the consistency of the estimators is obtained.
基金This project is supported by National Natural Science Foundation of China (N.59990470).
文摘The profile error evaluation of complex curves and surfaces expressed inparametric form is considered. The linear error model is established on the base of two hypothesesfirstly. Then the profile error evaluation is converted into one of these optimal formulations:MINIMAX, MAXMIN and MINIDEX problems, which are easier to be solved than the initial form. To eachone of them, geometric condition and algebraic condition are presented to arbitrate whether theideal element reaches to the optimal position. Exchange algorithm is proven highly effective insearching for solutions to these optimization problems. At last some key problems in tolerance offreeform surfaces and curves in B spline method are discussed.
基金supported by the National High Technology Research and Development Program of China (863 Program) (2007AA04Z227)
文摘This paper proposes a robust method of parameter estimation and data classification for multiple-structural data based on the linear error in variable（EIV） model.The traditional EIV model fitting problem is analyzed and a robust growing algorithm is developed to extract the underlying linear structure of the observed data.Under the structural density assumption,the C-step technique borrowed from the Rousseeuw＇s robust MCD estimator is used to keep the algorithm robust and the mean-shift algorithm is adopted to ensure a good initialization.To eliminate the model ambiguities of the multiple-structural data,statistical hypotheses tests are used to refine the data classification and improve the accuracy of the model parameter estimation.Experiments show that the efficiency and robustness of the proposed algorithm.
