Using the so-called martingale difference correlation(MDC), we propose a novel censoredconditional-quantile screening approach for ultrahigh-dimensional survival data with heterogeneity(which is often present in such ...Using the so-called martingale difference correlation(MDC), we propose a novel censoredconditional-quantile screening approach for ultrahigh-dimensional survival data with heterogeneity(which is often present in such data). By incorporating a weighting scheme, this method is a natural extension of MDCbased conditional quantile screening, as considered by Shao and Zhang(2014), to handle ultrahigh-dimensional survival data. The proposed screening procedure has a sure-screening property under certain technical conditions and an excellent capability of detecting the nonlinear relationship between independent and censored dependent variables. Both simulation results and an analysis of real data demonstrate the effectiveness of the new censored conditional quantile-screening procedure.展开更多
With the rapid-growth-in-size scientific data in various disciplines, feature screening plays an important role to reduce the high-dimensionality to a moderate scale in many scientific fields. In this paper, we introd...With the rapid-growth-in-size scientific data in various disciplines, feature screening plays an important role to reduce the high-dimensionality to a moderate scale in many scientific fields. In this paper, we introduce a unified and robust model-free feature screening approach for high-dimensional survival data with censoring, which has several advantages: it is a model-free approach under a general model framework, and hence avoids the complication to specify an actual model form with huge number of candidate variables; under mild conditions without requiring the existence of any moment of the response, it enjoys the ranking consistency and sure screening properties in ultra-high dimension. In particular, we impose a conditional independence assumption of the response and the censoring variable given each covariate, instead of assuming the censoring variable is independent of the response and the covariates. Moreover, we also propose a more robust variant to the new procedure, which possesses desirable theoretical properties without any finite moment condition of the predictors and the response. The computation of the newly proposed methods does not require any complicated numerical optimization and it is fast and easy to implement. Extensive numerical studies demonstrate that the proposed methods perform competitively for various configurations. Application is illustrated with an analysis of a genetic data set.展开更多
基金supported by the National Statistical Scientific Research Projects(Grant No.2015LZ54)
文摘Using the so-called martingale difference correlation(MDC), we propose a novel censoredconditional-quantile screening approach for ultrahigh-dimensional survival data with heterogeneity(which is often present in such data). By incorporating a weighting scheme, this method is a natural extension of MDCbased conditional quantile screening, as considered by Shao and Zhang(2014), to handle ultrahigh-dimensional survival data. The proposed screening procedure has a sure-screening property under certain technical conditions and an excellent capability of detecting the nonlinear relationship between independent and censored dependent variables. Both simulation results and an analysis of real data demonstrate the effectiveness of the new censored conditional quantile-screening procedure.
基金supported by the Research Grant Council of Hong Kong (Grant Nos. 509413 and 14311916)Direct Grants for Research of The Chinese University of Hong Kong (Grant Nos. 3132754 and 4053235)+3 种基金the Natural Science Foundation of Jiangxi Province (Grant No. 20161BAB201024)the Key Science Fund Project of Jiangxi Province Eduction Department (Grant No. GJJ150439)National Natural Science Foundation of China (Grant Nos. 11461029, 11601197 and 61562030)the Canadian Institutes of Health Research (Grant No. 145546)
文摘With the rapid-growth-in-size scientific data in various disciplines, feature screening plays an important role to reduce the high-dimensionality to a moderate scale in many scientific fields. In this paper, we introduce a unified and robust model-free feature screening approach for high-dimensional survival data with censoring, which has several advantages: it is a model-free approach under a general model framework, and hence avoids the complication to specify an actual model form with huge number of candidate variables; under mild conditions without requiring the existence of any moment of the response, it enjoys the ranking consistency and sure screening properties in ultra-high dimension. In particular, we impose a conditional independence assumption of the response and the censoring variable given each covariate, instead of assuming the censoring variable is independent of the response and the covariates. Moreover, we also propose a more robust variant to the new procedure, which possesses desirable theoretical properties without any finite moment condition of the predictors and the response. The computation of the newly proposed methods does not require any complicated numerical optimization and it is fast and easy to implement. Extensive numerical studies demonstrate that the proposed methods perform competitively for various configurations. Application is illustrated with an analysis of a genetic data set.
