ScholarBank@NUShttps://scholarbank.nus.edu.sgThe DSpace digital repository system captures, stores, indexes, preserves, and distributes digital research material.Wed, 08 Feb 2023 11:47:47 GMT2023-02-08T11:47:47Z5091- Edge Selection for Undirected Graphshttps://scholarbank.nus.edu.sg/handle/10635/148206Title: Edge Selection for Undirected Graphs
Authors: Meng Hwee Victor Ong; Sanjay Chaudhuri; Berwin Turlach
Fri, 31 Aug 2018 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1482062018-08-31T00:00:00Z
- Generalized linear models incorporating population level information: An empirical-likelihood-based approachhttps://scholarbank.nus.edu.sg/handle/10635/105160Title: Generalized linear models incorporating population level information: An empirical-likelihood-based approach
Authors: Chaudhuri, S.; Handcock, M.S.; Rendall, M.S.
Abstract: In many situations information from a sample of individuals can be supplemented by population level information on the relationship between a dependent variable and explanatory variables. Inclusion of the population level information can reduce bias and increase the efficiency of the parameter estimates. Population level information can be incorporated via constraints on functions of the model parameters. In general the constraints are non-linear, making the task of maximum likelihood estimation more difficult. We develop an alternative approach exploiting the notion of an empirical likelihood. It is shown that, within the framework of generalized linear models, the population level information corresponds to linear constraints, which are comparatively easy to handle. We provide a two-step algorithm that produces parameter estimates by using only unconstrained estimation. We also provide computable expressions for the standard errors. We give an application to demographic hazard modelling by combining panel survey data with birth registration data to estimate annual birth probabilities by parity. © 2008 Royal Statistical Society.
Tue, 01 Jan 2008 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1051602008-01-01T00:00:00Z
- Covariance based Moment Equations for Improved Variance Component Estimationhttps://scholarbank.nus.edu.sg/handle/10635/236366Title: Covariance based Moment Equations for Improved Variance Component Estimation
Authors: Sanjay Chaudhuri; Tatsuya Kubokawa; Shonosuke Sugasawa
Fri, 02 Dec 2022 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/2363662022-12-02T00:00:00Z
- Evaluation of hardness of the interfacial reaction products at the alumina-stainless steel brazed interface by modeling of nanoindentation resultshttps://scholarbank.nus.edu.sg/handle/10635/105136Title: Evaluation of hardness of the interfacial reaction products at the alumina-stainless steel brazed interface by modeling of nanoindentation results
Authors: Kar, A.; Chaudhuri, S.; Sen, P.K.; Ray, A.K.
Abstract: We have analyzed the 304 stainless steel (SS)-(Ag-Cu-Ti)-alumina brazed interface using scanning electron microscopy, electron probe microanalysis and nanoindentation. The SS interface exhibits increased bond strength and a larger diffusion zone compared to the alumina interface. In order to explain the nature of variation in hardness, we have fitted a second-degree Hermite polynomial-based model to the experimental observations of the nanoindentation results, across the reaction product zone of both interfaces. © 2007 Acta Materialia Inc.
Thu, 01 Nov 2007 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1051362007-11-01T00:00:00Z
- Qualitative inequalities for squared partial correlations of a Gaussian random vectorhttps://scholarbank.nus.edu.sg/handle/10635/125061Title: Qualitative inequalities for squared partial correlations of a Gaussian random vector
Authors: Chaudhuri, S.
Abstract: We describe various sets of conditional independence relationships, sufficient for qualitatively comparing non-vanishing squared partial correlations of a Gaussian random vector. These sufficient conditions are satisfied by several graphical Markov models. Rules for comparing degree of association among the vertices of such Gaussian graphical models are also developed. We apply these rules to compare conditional dependencies on Gaussian trees. In particular for trees, we show that such dependence can be completely characterised by the length of the paths joining the dependent vertices to each other and to the vertices conditioned on. We also apply our results to postulate rules for model selection for polytree models. Our rules apply to mutual information of Gaussian random vectors as well. © 2013 The Institute of Statistical Mathematics, Tokyo.
Wed, 01 Jan 2014 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1250612014-01-01T00:00:00Z
- Two step-down tests for equality of covariance matriceshttps://scholarbank.nus.edu.sg/handle/10635/105446Title: Two step-down tests for equality of covariance matrices
Authors: Chaudhuri, S.; Perlman, M.D.
Abstract: The classical problem of testing the equality of the covariance matrices from k ≥ 2 p-dimensional normal populations is reexamined. The likelihood ratio (LR) statistic, also called Bartlett's statistic, can be decomposed in two ways, corresponding to two distinct component-wise decompositions of the null hypothesis in terms of the covariance matrices or precision matrices, respectively. The factors of the LR statistic that appear in these two decompositions can be interpreted as conditional and unconditional LR statistics for the component-wise null hypotheses, and their mutual independence under the null hypothesis allows the determination of the overall significance level. © 2006 Elsevier Inc. All rights reserved.
Tue, 01 Aug 2006 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1054462006-08-01T00:00:00Z
- Consistent estimation of the minimum normal mean under the tree-order restrictionhttps://scholarbank.nus.edu.sg/handle/10635/105070Title: Consistent estimation of the minimum normal mean under the tree-order restriction
Authors: Chaudhuri, S.; Perlman, M.D.
Abstract: Let (Xij {divides} j = 1, ..., ni (s), i = 0, 1, ..., s) be independent observations from s + 1 univariate normal populations, with Xij ∼ N (μi, σ2). The tree-order restriction (μ0 ≤ μi, i = 1, ..., s) arises naturally when comparing a treatment (μ0) to several controls (μ1, ..., μs). When the sample sizes and population means and variances are equal and fixed, the maximum likelihood-based estimator (MLBE) of μ0 is negatively biased and diverges to - ∞ a.s. as s → ∞, leading some to assert that maximum likelihood may "fail disastrously" in order-restricted estimation. By viewing this problem as one of estimating a target parameter μ0 in the presence of an increasing number of nuisance parameters μ1, ..., μs, however, this behavior is reminiscent of the classical Neyman-Scott example. This suggests an alternative formulation of the problem wherein the sample size n0 (s) for the target parameter increases with s. Here the MLBE of μ0 is either consistent or admits a bias-reducing adjustment, depending on the rate of increase of n0 (s). The consistency of an estimator due to Cohen and Sackrowitz [2002. Inference for the model of several treatments and a control. J. Statist. Plann. Inference 107, 89-101] is also discussed. © 2007 Elsevier B.V. All rights reserved.
Thu, 01 Nov 2007 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1050702007-11-01T00:00:00Z
- Empirical likelihood for small area estimationhttps://scholarbank.nus.edu.sg/handle/10635/105118Title: Empirical likelihood for small area estimation
Authors: Chaudhuri, S.; Ghosh, M.
Abstract: Current methodologies in small area estimation are mostly either parametric or heavily dependent on the assumed linearity of the estimators of the small area means. We discuss an alternative empirical likelihood-based Bayesian approach, which neither requires a parametric likelihood nor assumes linearity of the estimators, and can handle both discrete and continuous data in a unified manner. Empirical likelihoods for both area- and unit-level models are introduced. We discuss the suitability of the proposed likelihoods in Bayesian inference and illustrate their performances on a real dataset and a simulation study. © 2011 Biometrika Trust.
Wed, 01 Jun 2011 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1051182011-06-01T00:00:00Z
- Estimation of a covariance matrix with zeroshttps://scholarbank.nus.edu.sg/handle/10635/105131Title: Estimation of a covariance matrix with zeros
Authors: Chaudhuri, S.; Drton, M.; Richardson, T.S.
Abstract: We consider estimation of the covariance matrix of a multivariate random vector under the constraint that certain covariances are zero. We first present an algorithm, which we call iterative conditional fitting, for computing the maximum likelihood estimate of the constrained covariance matrix, under the assumption of multivariate normality. In contrast to previous approaches, this algorithm has guaranteed convergence properties. Dropping the assumption of multivariate normality, we show how to estimate the covariance matrix in an empirical likelihood approach. These approaches are then compared via simulation and on an example of gene expression. © 2007 Biometrika Trust.
Thu, 01 Mar 2007 00:00:00 GMThttps://scholarbank.nus.edu.sg/handle/10635/1051312007-03-01T00:00:00Z