In this paper, we propose a new two-sample distribution-free procedure for testing group-by-time interaction effect in repeated measurements from a linear mixed model setting. The test statistic is based on the maximum difference of partial sums (MDPS) over time points between the two groups. Although the test has a biomedical focus, it can be applied in fields that the study is designed and monitored to be balanced and complete with equal sample sizes as would be generally done in a controlled experiment. The asymptotic null distribution of the test statistic was also derived based on the maxima of Brownian bridge under two different conditions. The simulations revealed that MDPS performed markedly better than the commonly used unstructured multivariate approach (UMA) to profile analysis. However, the empirical powers of MDPS test were convincingly close to those of the best-fitting linear mixed model (LMM).
{"title":"A distribution-free test of parallelism for two-sample repeated measurements","authors":"Mehrdad Vossoughi , S.M.T. Ayatollahi , Mina Towhidi , Seyyed Taghi Heydari","doi":"10.1016/j.stamet.2015.12.001","DOIUrl":"10.1016/j.stamet.2015.12.001","url":null,"abstract":"<div><p><span>In this paper, we propose a new two-sample distribution-free procedure for testing group-by-time interaction effect in repeated measurements from a linear mixed model setting. The test statistic is based on the maximum difference of partial sums (MDPS) over </span>time points<span> between the two groups. Although the test has a biomedical focus, it can be applied in fields that the study is designed and monitored to be balanced and complete with equal sample sizes as would be generally done in a controlled experiment. The asymptotic null<span> distribution of the test statistic was also derived based on the maxima of Brownian bridge under two different conditions. The simulations revealed that MDPS performed markedly better than the commonly used unstructured multivariate approach (UMA) to profile analysis. However, the empirical powers of MDPS test were convincingly close to those of the best-fitting linear mixed model (LMM).</span></span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.12.001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1016/j.stamet.2015.11.001
Afshin Almasi , Mohammad Reza Eshraghian , Abbas Moghimbeigi , Abbas Rahimi , Kazem Mohammad , Sadegh Fallahigilan
Poisson or zero-inflated Poisson models often fail to fit count data either because of over- or underdispersion relative to the Poisson distribution. Moreover, data may be correlated due to the hierarchical study design or the data collection methods. In this study, we propose a multilevel zero-inflated generalized Poisson regression model that can address both over- and underdispersed count data. Random effects are assumed to be independent and normally distributed. The method of parameter estimation is EM algorithm base on expectation and maximization which falls into the general framework of maximum-likelihood estimations. The performance of the approach was illustrated by data regarding an index of tooth caries on 9-year-old children. Using various dispersion parameters, through Monte Carlo simulations, the multilevel ZIGP yielded more accurate parameter estimates, especially for underdispersed data.
{"title":"Multilevel zero-inflated Generalized Poisson regression modeling for dispersed correlated count data","authors":"Afshin Almasi , Mohammad Reza Eshraghian , Abbas Moghimbeigi , Abbas Rahimi , Kazem Mohammad , Sadegh Fallahigilan","doi":"10.1016/j.stamet.2015.11.001","DOIUrl":"10.1016/j.stamet.2015.11.001","url":null,"abstract":"<div><p><span><span><span>Poisson or zero-inflated Poisson models often fail to fit count data either because of over- or underdispersion relative to the </span>Poisson distribution. Moreover, data may be correlated due to the hierarchical study design or the data collection methods. In this study, we propose a multilevel zero-inflated generalized Poisson regression model that can address both over- and underdispersed count data. Random effects are assumed to be independent and normally distributed. The method of parameter estimation is </span>EM algorithm base on expectation and maximization which falls into the general framework of maximum-likelihood estimations. The performance of the approach was illustrated by data regarding an index of tooth caries on 9-year-old children. Using various </span>dispersion parameters<span>, through Monte Carlo simulations, the multilevel ZIGP yielded more accurate parameter estimates, especially for underdispersed data.</span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.11.001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-05-01DOI: 10.1016/j.stamet.2015.11.002
Pao-sheng Shen
We analyze doubly truncated data using semiparametric transformation models. It is demonstrated that the extended estimating equations of Cheng et al. (1995) can be used to analyze doubly truncated data. The asymptotic properties of the proposed estimators are derived. A simulation study is conducted to investigate the performance of the proposed estimators.
我们用半参数变换模型分析双截断数据。证明了Cheng et al.(1995)的扩展估计方程可以用于分析双截断数据。给出了所提估计量的渐近性质。仿真研究了所提出的估计器的性能。
{"title":"Analysis of transformation models with doubly truncated data","authors":"Pao-sheng Shen","doi":"10.1016/j.stamet.2015.11.002","DOIUrl":"10.1016/j.stamet.2015.11.002","url":null,"abstract":"<div><p><span>We analyze doubly truncated data using semiparametric transformation models. It is demonstrated that the extended estimating equations of Cheng et al. (1995) can be used to analyze doubly truncated data. The </span>asymptotic properties of the proposed estimators are derived. A simulation study is conducted to investigate the performance of the proposed estimators.</p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.11.002","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-03-01DOI: 10.1016/j.stamet.2015.08.001
W. Kössler , Narinder Kumar
In the usual two-sample scale problem it is assumed that the two populations have a common median. We consider the case where the common quantile may be other than a half. We investigate a quite general class, all members are based on -statistics where the minima and maxima of subsamples of various sizes are used. The asymptotic efficacies are investigated in detail. We construct an adaptive test where all statistics involved are suitably chosen. It is shown that the proposed adaptive test has good asymptotic and finite power properties.
{"title":"An adaptive test for the two-sample scale problem where the common quantile may be different from the median","authors":"W. Kössler , Narinder Kumar","doi":"10.1016/j.stamet.2015.08.001","DOIUrl":"10.1016/j.stamet.2015.08.001","url":null,"abstract":"<div><p><span>In the usual two-sample scale problem it is assumed that the two populations have a common median. We consider the case where the common quantile may be other than a half. We investigate a quite general class, all members are based on </span><span><math><mi>U</mi></math></span><span>-statistics where the minima and maxima of subsamples of various sizes are used. The asymptotic efficacies are investigated in detail. We construct an adaptive test where all statistics involved are suitably chosen. It is shown that the proposed adaptive test has good asymptotic and finite power properties.</span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.08.001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In competing risks data, missing failure types (causes) is a very common phenomenon. In a general missing pattern, if a failure type is not observed, one observes a set of possible types containing the true type along with the failure time. Dewanji and Sengupta (2003) considered nonparametric estimation of the cause-specific hazard rates and suggested a Nelson–Aalen type estimator under such general missing pattern. In this work, we deal with the regression problem, in which the cause-specific hazard rates may depend on some covariates, and consider estimation of the regression coefficients and the cause-specific baseline hazards under the general missing pattern using some semi-parametric models. We consider two different proportional hazards type semi-parametric models for our analysis. Simulation studies from both the models are carried out to investigate the finite sample properties of the estimators. We also consider an example from an animal experiment to illustrate our methodology.
{"title":"Regression analysis of competing risks data with general missing pattern in failure types","authors":"Anup Dewanji , P.G. Sankaran , Debasis Sengupta , Bappa Karmakar","doi":"10.1016/j.stamet.2015.09.002","DOIUrl":"10.1016/j.stamet.2015.09.002","url":null,"abstract":"<div><p><span><span>In competing risks data, missing failure types (causes) is a very common phenomenon. In a general missing pattern, if a failure type is not observed, one observes a set of possible types containing the true type along with the failure time. Dewanji and Sengupta (2003) considered nonparametric estimation of the cause-specific hazard rates and suggested a Nelson–Aalen </span>type estimator under such general missing pattern. In this work, we deal with the regression problem, in which the cause-specific hazard rates may depend on some </span>covariates<span>, and consider estimation of the regression coefficients and the cause-specific baseline hazards under the general missing pattern using some semi-parametric models. We consider two different proportional hazards type semi-parametric models for our analysis. Simulation studies from both the models are carried out to investigate the finite sample properties of the estimators. We also consider an example from an animal experiment to illustrate our methodology.</span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.09.002","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-03-01DOI: 10.1016/j.stamet.2015.10.003
Julian Górny, Erhard Cramer
Generalized Type-I and Type-II hybrid censoring schemes as proposed in Chandrasekar et al. (2004) are extended to progressively Type-II censored data. Using the spacings’ based approach due to Cramer and Balakrishnan (2013), we obtain explicit expressions for the density functions of the MLEs. The resulting formulas are given in terms of B-spline functions so that they can be easily and efficiently implemented on a computer.
{"title":"Exact likelihood inference for exponential distributions under generalized progressive hybrid censoring schemes","authors":"Julian Górny, Erhard Cramer","doi":"10.1016/j.stamet.2015.10.003","DOIUrl":"10.1016/j.stamet.2015.10.003","url":null,"abstract":"<div><p><span>Generalized Type-I and Type-II hybrid censoring schemes as proposed in Chandrasekar et al. (2004) are extended to progressively Type-II </span>censored data<span>. Using the spacings’ based approach due to Cramer and Balakrishnan (2013), we obtain explicit expressions for the density functions of the MLEs. The resulting formulas are given in terms of B-spline functions so that they can be easily and efficiently implemented on a computer.</span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.10.003","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-03-01DOI: 10.1016/j.stamet.2015.10.001
M. Kayid , S. Izadkhah , H. Alhalees
The purposes of this paper are to introduce a new stochastic order and to study its reliability properties. Some characterizations and preservation properties of the new order under reliability operations of monotone transformation, mixture, weighted distributions and shock models are discussed. In addition, a new class of life distributions is proposed, and some of its reliability properties are investigated. Finally, to illustrate the concepts, some applications in the context of reliability theory and life testing are presented.
{"title":"Combination of mean residual life order with reliability applications","authors":"M. Kayid , S. Izadkhah , H. Alhalees","doi":"10.1016/j.stamet.2015.10.001","DOIUrl":"10.1016/j.stamet.2015.10.001","url":null,"abstract":"<div><p>The purposes of this paper are to introduce a new stochastic order<span> and to study its reliability properties. Some characterizations and preservation properties of the new order under reliability operations of monotone transformation, mixture, weighted distributions and shock models are discussed. In addition, a new class of life distributions is proposed, and some of its reliability properties are investigated. Finally, to illustrate the concepts, some applications in the context of reliability theory and life testing are presented.</span></p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.10.001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120871445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-03-01DOI: 10.1016/j.stamet.2015.09.001
Gwo Dong Lin , Chin-Diew Lai , K. Govindaraju
We first review the basic properties of Marshall–Olkin bivariate exponential distribution (BVE) and then investigate its correlation structure. We provide the correct reasonings for deriving some properties of the Marshall–Olkin BVE and show that the correlation of the BVE is always smaller than that of its copula regardless of the parameters. The latter implies that the BVE does not have Lancaster’s phenomenon (any nonlinear transformation of variables decreases the correlation in absolute value). The dependence structure of the BVE is also investigated.
{"title":"Correlation structure of the Marshall–Olkin bivariate exponential distribution","authors":"Gwo Dong Lin , Chin-Diew Lai , K. Govindaraju","doi":"10.1016/j.stamet.2015.09.001","DOIUrl":"10.1016/j.stamet.2015.09.001","url":null,"abstract":"<div><p><span><span>We first review the basic properties of Marshall–Olkin bivariate </span>exponential distribution<span> (BVE) and then investigate its correlation structure<span>. We provide the correct reasonings for deriving some properties of the Marshall–Olkin BVE and show that the correlation of the BVE is always smaller than that of its copula regardless of the parameters. The latter implies that the BVE does not have Lancaster’s phenomenon (any nonlinear transformation of variables decreases the correlation in absolute value). The </span></span></span>dependence structure of the BVE is also investigated.</p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.09.001","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-03-01DOI: 10.1016/j.stamet.2015.10.002
Athanasios C. Rakitzis , Philippe Castagliola , Petros E. Maravelakis
In this work, we propose and study a two-parameter modification of the ordinary Poisson distribution that is suitable for the modeling of non-typical count data. This model can be viewed as an extension of the zero-inflated Poisson distribution. We derive the proposed model as a special case of a general one and focus our study on it. The theoretical properties for each model are given, while estimation methods for the two-parameter model are discussed as well. Three practical examples illustrate its usefulness. The results show that the proposed model is very flexible in the modeling of various types of count data.
{"title":"A two-parameter general inflated Poisson distribution: Properties and applications","authors":"Athanasios C. Rakitzis , Philippe Castagliola , Petros E. Maravelakis","doi":"10.1016/j.stamet.2015.10.002","DOIUrl":"10.1016/j.stamet.2015.10.002","url":null,"abstract":"<div><p>In this work, we propose and study a two-parameter modification of the ordinary Poisson distribution that is suitable for the modeling of non-typical count data. This model can be viewed as an extension of the zero-inflated Poisson distribution. We derive the proposed model as a special case of a general one and focus our study on it. The theoretical properties for each model are given, while estimation methods for the two-parameter model are discussed as well. Three practical examples illustrate its usefulness. The results show that the proposed model is very flexible in the modeling of various types of count data.</p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.10.002","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-01-01DOI: 10.1016/j.stamet.2015.07.004
Kaushik Mahata , Amit Mitra , Sharmishtha Mitra
In this paper, we consider the problem of robust M-estimation of parameters of nonlinear signal processing models. We investigate the conditions under which estimators are strongly consistent for convex and non-convex penalty functions and a wide class of noise scenarios, contaminating the actual transmitted signal. It is shown that the M-estimators of a general nonlinear signal model are asymptotically consistent with probability one under different sets of sufficient conditions on loss function and noise distribution. Simulations are performed for nonlinear superimposed sinusoidal model to observe the small sample performance of the M-estimators for various heavy tailed error distributions, outlier contamination levels and sample sizes.
{"title":"Consistency of M-estimators of nonlinear signal processing models","authors":"Kaushik Mahata , Amit Mitra , Sharmishtha Mitra","doi":"10.1016/j.stamet.2015.07.004","DOIUrl":"10.1016/j.stamet.2015.07.004","url":null,"abstract":"<div><p>In this paper, we consider the problem of robust M-estimation of parameters of nonlinear signal processing models. We investigate the conditions under which estimators are strongly consistent for convex and non-convex penalty functions and a wide class of noise scenarios, contaminating the actual transmitted signal. It is shown that the M-estimators of a general nonlinear signal model are asymptotically consistent with probability one under different sets of sufficient conditions on loss function and noise distribution. Simulations are performed for nonlinear superimposed sinusoidal model to observe the small sample performance of the M-estimators for various heavy tailed error distributions, outlier contamination levels and sample sizes.</p></div>","PeriodicalId":48877,"journal":{"name":"Statistical Methodology","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/j.stamet.2015.07.004","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"55093100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}