Showing papers by "Pranab Kumar Sen published in 2006"
TL;DR: In this paper, the robust estimation of a (linear) regression parameter (vector) in the presence of a nuisance scale parameter is considered when it is a priori suspected that the regression could be r...
Abstract: The problem of robust estimation of a (linear) regression parameter (vector), in the presence of nuisance scale parameter, is considered when it is a priori suspected that the regression could be r...
42 citations
TL;DR: In this article, Roy's union-intersection principle is used to solve the problem of high-dimensional (K) low sample size (n) environments, where inequality, order or general shape constraints crop up in complex ways, and as a result, likelihood based optimal statistical inference proce- dures may not exist, at least, may not be in manageable form.
Abstract: In high-dimension (K) low sample size (n) environments, often nonlinear, inequality, order or general shape constraints crop up in complex ways, and as a result, likelihood based optimal statistical inference proce- dures may not exist, at least, may not be in manageable form. While some of these inference problems can be treated in asymptotic setups, the curse of dimensionality (i.e., K >> n with often n small) calls for a different type of asymptotics (in K) with different perspectives. Roy's union-intersection principle provides some alternative approaches, generally more amenable for K >> n environments. This scenario is appraised with two important sta- tistical problems in genomic studies: a large number of (possibly dependent) genes with heterogeneity amidst a smaller sample create impasses for stan- dard robust inference. These perspectives are examined here in a nonstandard statistical analysis.
25 citations
Journal Article•
TL;DR: In this article, a semi-parametric transformation model is proposed for failure time data from case-cohort studies, where the covariates are assembled only for a randomly selected sub-co-hort from the entire cohort and additional cases outside the subco-hoort.
Abstract: Semiparametric transformation models are considered for failure time data from case-cohort studies, where the covariates are assembled only for a randomly selected subcohort from the entire cohort and additional cases outside the subcohort. We present the estimating procedures for the regression parameters and survival probability. The asymptotic properties of the resulting estimators are developed based on asymptotic results for U-statistics, martingales, stochastic processes and finite population sampling.
13 citations
TL;DR: The proposed fully operational parametric statistical model is fortified with flexibility to withstand use in several organisms and adaptability to intersite dependence, and the problem of homogeneity among groups of genomic sequences incorporating the available evidence of diversity in categorical data models is considered.
Abstract: Genomic data models relate to a large number of positions exhibiting categorical responses, and larger part of that provides no statistical information . For this reason, a reduction in complexity ...
2 citations
TL;DR: In this article, the interplay of robustness, admissibility and shrinkage phenomenon in general multivariate location models (not necessarily elliptically or spherically symmetric) is illustrated and applied to Huber-type contamination models.
Abstract: Estimators of multivariate location parameters are generally dominated, in finite as well as asymptotic setups, by suitable shrinkage versions, and hence are inadmissible; such shrinkage estimators may not be admissible either. This feature is shared by maximum likelihood and many robust estimators. The interplay of robustness, admissibility and shrinkage phenomenon in some general multivariate location models (not necessarily elliptically or spherically symmetric) is illustrated and applied to Huber-type contamination models.
2 citations
TL;DR: In this article, the Fisher information is intricately linked to the asymptotic optimality of maximum likelihood estimators for parametric complete-data models, and it is shown that information in a single observation is well-defined and plays the same role as in the complete data model in characterizing the first-order optimality properties of associated estimators; computational aspects are also thoroughly appraised.
1 citations
TL;DR: In this paper, two robust R-estimators based on signed-rank statistics were proposed for the common location parameter, and their properties were studied when large heteroscedasticity is present or the distribution of random effect is abnormal.
Abstract: We consider multi-center experiments (for determining a consensus value) conducted in possibly heterogeneous set-ups leading to unbalanced heteroscedastic one-way random effects models. When normality of both the random components and their homoscedasticity are in doubt, standard statistical methods may not be valid. Two robust R-estimators (for the common location parameter), based on signed-rank statistics, are proposed and their properties studied. When large heteroscedasticity is present or the distribution of random effect is abnormal, the proposed estimators perform better than the classical weighted least squares and selected estimators. This feature is illustrated with an arsenic in oyster tissue problem, along with some other simulation studies.
1 citations