Exclusive hypothesis testing is a new and special class of hypothesis testing.This kind of testing can be applied in survival analysis to understand the association between genomics information and clinical informatio...Exclusive hypothesis testing is a new and special class of hypothesis testing.This kind of testing can be applied in survival analysis to understand the association between genomics information and clinical information about the survival time.Besides,it is well known that Cox's proportional hazards model is the most commonly used model for regression analysis of failure time.In this paper,the authors consider doing the exclusive hypothesis testing for Cox's proportional hazards model with right-censored data.The authors propose the comprehensive test statistics to make decision,and show that the corresponding decision rule can control the asymptotic TypeⅠerrors and have good powers in theory.The numerical studies indicate that the proposed approach works well for practical situations and it is applied to a set of real data arising from Rotterdam Breast Cancer Data study that motivated this study.展开更多
For partial linear model Y = X τ β 0 + g 0(T) + ∈ with unknown β 0 ∈ ? d and an unknown smooth function g 0, this paper considers the Huber-Dutter estimators of β 0, scale σ for the errors and the function g 0 ...For partial linear model Y = X τ β 0 + g 0(T) + ∈ with unknown β 0 ∈ ? d and an unknown smooth function g 0, this paper considers the Huber-Dutter estimators of β 0, scale σ for the errors and the function g 0 approximated by the smoothing B-spline functions, respectively. Under some regularity conditions, the Huber-Dutter estimators of β 0 and σ are shown to be asymptotically normal with the rate of convergence n ?1/2 and the B-spline Huber-Dutter estimator of g 0 achieves the optimal rate of convergence in nonparametric regression. A simulation study and two examples demonstrate that the Huber-Dutter estimator of β 0 is competitive with its M-estimator without scale parameter and the ordinary least square estimator.展开更多
We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estim...We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estimation of the covariate effect. The proposed estimator is asymptotically normal. Simulation studies are presented to show that the proposed method performs well with finite samples, and the proposed method is applied to a real data set.展开更多
To test genetic pleiotropy,the main difficulty lies in the failure to find a test statistic and calculate its p-value for determining whether to reject the null hypothesis or not.To deal with this issue,the authors pr...To test genetic pleiotropy,the main difficulty lies in the failure to find a test statistic and calculate its p-value for determining whether to reject the null hypothesis or not.To deal with this issue,the authors propose a quasi p-value,which plays the similar role as the usual p-value in genetic pleiotropy test.In the formula of the quasi p-value,the main task is to determine the weights.In this paper,the authors present two weighted methods based on the Bayesian rule and extend the proposed methods to study a single binary trait using a data-driven EM algorithm.Extensive simulation studies are conducted for the assessment of the two proposed methods and illustrate that the proposed methods improve the performance of power by comparing with the two-stage method.In addition,the authors apply the proposed methods to the data of tropical storms that occurred on the mainland of China since 1949,investigating the relationship between the landing site and predictive features of tropical storms,and showing that the landing site has a large influence on at least two features of typhoon.展开更多
基金supported by the National Natural Science Foundation of China under Grant Nos.11971064,12371262,and 12171374。
文摘Exclusive hypothesis testing is a new and special class of hypothesis testing.This kind of testing can be applied in survival analysis to understand the association between genomics information and clinical information about the survival time.Besides,it is well known that Cox's proportional hazards model is the most commonly used model for regression analysis of failure time.In this paper,the authors consider doing the exclusive hypothesis testing for Cox's proportional hazards model with right-censored data.The authors propose the comprehensive test statistics to make decision,and show that the corresponding decision rule can control the asymptotic TypeⅠerrors and have good powers in theory.The numerical studies indicate that the proposed approach works well for practical situations and it is applied to a set of real data arising from Rotterdam Breast Cancer Data study that motivated this study.
基金the National Natural Science Foundation of China (Grant Nos. 10671106, 10771017)
文摘For partial linear model Y = X τ β 0 + g 0(T) + ∈ with unknown β 0 ∈ ? d and an unknown smooth function g 0, this paper considers the Huber-Dutter estimators of β 0, scale σ for the errors and the function g 0 approximated by the smoothing B-spline functions, respectively. Under some regularity conditions, the Huber-Dutter estimators of β 0 and σ are shown to be asymptotically normal with the rate of convergence n ?1/2 and the B-spline Huber-Dutter estimator of g 0 achieves the optimal rate of convergence in nonparametric regression. A simulation study and two examples demonstrate that the Huber-Dutter estimator of β 0 is competitive with its M-estimator without scale parameter and the ordinary least square estimator.
基金supported by National Natural Science Foundation of China(Grant Nos.10901020 and 11371062)the Fundamental Research Funds for the Central Universities,Beijing Center for Mathematics and Information Interdisciplinary Sciences,China Zhongdian Project(Grant No.11131002)
文摘We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estimation of the covariate effect. The proposed estimator is asymptotically normal. Simulation studies are presented to show that the proposed method performs well with finite samples, and the proposed method is applied to a real data set.
基金supported by National Key Research and Development Program of China under Grant No.2017YFA0604903the National Natural Science Foundation of China under Grant No.11971064the Ph.D.start-up fund of Hubei University of Science and Technology under Grant No.BK201813。
文摘To test genetic pleiotropy,the main difficulty lies in the failure to find a test statistic and calculate its p-value for determining whether to reject the null hypothesis or not.To deal with this issue,the authors propose a quasi p-value,which plays the similar role as the usual p-value in genetic pleiotropy test.In the formula of the quasi p-value,the main task is to determine the weights.In this paper,the authors present two weighted methods based on the Bayesian rule and extend the proposed methods to study a single binary trait using a data-driven EM algorithm.Extensive simulation studies are conducted for the assessment of the two proposed methods and illustrate that the proposed methods improve the performance of power by comparing with the two-stage method.In addition,the authors apply the proposed methods to the data of tropical storms that occurred on the mainland of China since 1949,investigating the relationship between the landing site and predictive features of tropical storms,and showing that the landing site has a large influence on at least two features of typhoon.
