Pub Date : 2006-12-01DOI: 10.1177/1471082006071851
C. Faes, M. Aerts, H. Geys, L. Bijnens, L. Ver Donck, W. Lammers
Mixed models can be applied in a wide range of settings. Probably, they are most commonly used to handle grouping in the data. In addition, mixed models can be used for smoothing purposes as well. When dealing with non-normal data, the use of smoothing methods within the generalized linear mixed models (GLMM) framework is less familiar. We explore the use of GLMM for smoothing purposes in both spatial and longitudinal dimensions. The methodology is illustrated by analysis of spike potentials in the small intestine of different cats. Spatio-temporal models that use two-dimensional smoothing splines across the spatial dimension and random effects to account for the correlations during successive slow-waves are developed. A major advantage of the mixed-model approach is that it can handle smoothing together with grouping (or other types of correlations) in a unified model. In this way, areas with high spike incidence compared with other areas can be detected. Also, the temporal and spatial characteristics of spikes during successive slow-waves can be identified.
{"title":"GLMM approach to study the spatial and temporal evolution of spikes in the small intestine","authors":"C. Faes, M. Aerts, H. Geys, L. Bijnens, L. Ver Donck, W. Lammers","doi":"10.1177/1471082006071851","DOIUrl":"https://doi.org/10.1177/1471082006071851","url":null,"abstract":"Mixed models can be applied in a wide range of settings. Probably, they are most commonly used to handle grouping in the data. In addition, mixed models can be used for smoothing purposes as well. When dealing with non-normal data, the use of smoothing methods within the generalized linear mixed models (GLMM) framework is less familiar. We explore the use of GLMM for smoothing purposes in both spatial and longitudinal dimensions. The methodology is illustrated by analysis of spike potentials in the small intestine of different cats. Spatio-temporal models that use two-dimensional smoothing splines across the spatial dimension and random effects to account for the correlations during successive slow-waves are developed. A major advantage of the mixed-model approach is that it can handle smoothing together with grouping (or other types of correlations) in a unified model. In this way, areas with high spike incidence compared with other areas can be detected. Also, the temporal and spatial characteristics of spikes during successive slow-waves can be identified.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126784337","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 : 2006-12-01DOI: 10.1177/1471082006071848
A. Bowman
There is a wide variety of problems where the object of primary interest is a surface. Environmental studies in particular, where data often have a spatial structure, provide many examples where estimation of a surface is a central component of analysis. In these settings, the surfaces are often not well described by simple parametric models. Nonparametric regression therefore offers a convenient means of constructing surface estimates in a straightforward manner. In this paper, the issues associated with comparing such regression surfaces across different groups of data are discussed. Formal methods for assessing the equality of a collection of surfaces, or the suitability of a set of parallel surfaces, are described. These not only extend existing methods of nonparametric analysis of covariance but also allow the commonly occurring case of correlated errors to be incorporated. Graphical methods to provide insight into the sources of departure from a candidate model are also proposed. Several applications are provided to illustrate and explore the proposals.
{"title":"Comparing nonparametric surfaces","authors":"A. Bowman","doi":"10.1177/1471082006071848","DOIUrl":"https://doi.org/10.1177/1471082006071848","url":null,"abstract":"There is a wide variety of problems where the object of primary interest is a surface. Environmental studies in particular, where data often have a spatial structure, provide many examples where estimation of a surface is a central component of analysis. In these settings, the surfaces are often not well described by simple parametric models. Nonparametric regression therefore offers a convenient means of constructing surface estimates in a straightforward manner. In this paper, the issues associated with comparing such regression surfaces across different groups of data are discussed. Formal methods for assessing the equality of a collection of surfaces, or the suitability of a set of parallel surfaces, are described. These not only extend existing methods of nonparametric analysis of covariance but also allow the commonly occurring case of correlated errors to be incorporated. Graphical methods to provide insight into the sources of departure from a candidate model are also proposed. Several applications are provided to illustrate and explore the proposals.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129623103","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 : 2006-12-01DOI: 10.1177/1471082006071849
M. Geraci, M. Bottai
We propose a method for reducing the error of the prediction of a quantity of interest when the outcome has missing values that are suspected to be nonignorable and the data are correlated in space. We develop a maximum likelihood approach for the parameter estimation of semi-parametric regressions in a mixed model framework. We apply the proposed method to phytoplankton data collected at fixed stations in the Chesapeake Bay, for which chlorophyll data coming from remote sensing are available. A simulation study is also performed. The availability of a variable correlated to the response allows us to achieve a substantial reduction of the prediction error of the expected value of the smoother, without having to specify a nonignorable model.
{"title":"Use of auxiliary data in semi-parametric spatial regression with nonignorable missing responses","authors":"M. Geraci, M. Bottai","doi":"10.1177/1471082006071849","DOIUrl":"https://doi.org/10.1177/1471082006071849","url":null,"abstract":"We propose a method for reducing the error of the prediction of a quantity of interest when the outcome has missing values that are suspected to be nonignorable and the data are correlated in space. We develop a maximum likelihood approach for the parameter estimation of semi-parametric regressions in a mixed model framework. We apply the proposed method to phytoplankton data collected at fixed stations in the Chesapeake Bay, for which chlorophyll data coming from remote sensing are available. A simulation study is also performed. The availability of a variable correlated to the response allows us to achieve a substantial reduction of the prediction error of the expected value of the smoother, without having to specify a nonignorable model.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124064676","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 : 2006-12-01DOI: 10.1177/1471082006071841
Li Zhang, B. Mukherjee, M. Ghosh, R. Wu
A two-stage parametric Bayesian method is proposed to examine the association between a candidate gene and the occurrence of a disease after accounting for population substructure. This procedure, implemented via a Markov chain Monte Carlo numerical integration technique, first estimates the posterior probability of different unknown population substructures and then integrates this information into a disease-gene association model through the technique of Bayesian model averaging. The model relaxes certain assumptions of previous analyses and provides a unified computational framework to obtain an estimate of the log odds ratio parameter corresponding to the genetic factor after allowing for the allele frequencies to vary across subpopulations. The uncertainty in estimating the population substructure is taken into account while providing credible intervals for parameters in the disease-gene association model. Simulations on unmatched case-control studies that mimic an admixed Argentinean population are performed to demonstrate the statistical properties of our model. The method is also applied to a real data set coming from a genetic association study on obesity.
{"title":"Bayesian modeling for genetic association in case-control studies: accounting for unknown population substructure","authors":"Li Zhang, B. Mukherjee, M. Ghosh, R. Wu","doi":"10.1177/1471082006071841","DOIUrl":"https://doi.org/10.1177/1471082006071841","url":null,"abstract":"A two-stage parametric Bayesian method is proposed to examine the association between a candidate gene and the occurrence of a disease after accounting for population substructure. This procedure, implemented via a Markov chain Monte Carlo numerical integration technique, first estimates the posterior probability of different unknown population substructures and then integrates this information into a disease-gene association model through the technique of Bayesian model averaging. The model relaxes certain assumptions of previous analyses and provides a unified computational framework to obtain an estimate of the log odds ratio parameter corresponding to the genetic factor after allowing for the allele frequencies to vary across subpopulations. The uncertainty in estimating the population substructure is taken into account while providing credible intervals for parameters in the disease-gene association model. Simulations on unmatched case-control studies that mimic an admixed Argentinean population are performed to demonstrate the statistical properties of our model. The method is also applied to a real data set coming from a genetic association study on obesity.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125566753","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 : 2006-12-01DOI: 10.1177/1471082006071844
S. Cecere, A. Jara, E. Lesaffre
Based on a data set obtained in a large dental longitudinal study, conducted in Flanders (Belgium), the joint emergence distribution of seven teeth was modeled as a function of gender and caries experience on primary teeth. Besides establishing the marginal dependence of emergence on the covariates, there was also interest in examining the impact of the covariates on the association among emergence times. This allows the establishment of the preferred rankings of emergence and their dependence on covariates. To this end, the covariance matrix was modeled as a function of covariates. Modeling the covariance matrix in this way needs to ensure the positive definiteness of the covariance matrix and it is preferable that the regression parameters of the model are interpretable. The modified Cholesky decomposition of the covariance matrix, as suggested by Pourahmadi, splits up the covariance matrix into two parts where the parameters can be interpreted, given a natural ranking of the responses. This approach was used here taking into account that the emergence times are interval-censored. Hence, we opted for a Bayesian implementation of the data augmentation algorithm.
{"title":"Analyzing the emergence times of permanent teeth: an example of modeling the covariance matrix with interval-censored data","authors":"S. Cecere, A. Jara, E. Lesaffre","doi":"10.1177/1471082006071844","DOIUrl":"https://doi.org/10.1177/1471082006071844","url":null,"abstract":"Based on a data set obtained in a large dental longitudinal study, conducted in Flanders (Belgium), the joint emergence distribution of seven teeth was modeled as a function of gender and caries experience on primary teeth. Besides establishing the marginal dependence of emergence on the covariates, there was also interest in examining the impact of the covariates on the association among emergence times. This allows the establishment of the preferred rankings of emergence and their dependence on covariates. To this end, the covariance matrix was modeled as a function of covariates. Modeling the covariance matrix in this way needs to ensure the positive definiteness of the covariance matrix and it is preferable that the regression parameters of the model are interpretable. The modified Cholesky decomposition of the covariance matrix, as suggested by Pourahmadi, splits up the covariance matrix into two parts where the parameters can be interpreted, given a natural ranking of the responses. This approach was used here taking into account that the emergence times are interval-censored. Hence, we opted for a Bayesian implementation of the data augmentation algorithm.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131686432","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 : 2006-10-01DOI: 10.1191/1471082X06st116oa
Kahadawala Cooray
A three-parameter generalization of the Weibull distribution is presented to deal with general situations in modeling survival process with various shapes in the hazard function. This generalized Weibull distribution will be referred to as the odd Weibull family, as it is derived by considering the distributions of the odds of the Weibull and inverse Weibull families. As a result, the odd Weibull family is not only useful for testing goodness-of-fit of the Weibull and inverse Weibull as submodels, but it is also convenient for modeling and fitting different data sets, especially in the presence of censoring. The model parameters for uncensored data are estimated in two different ways because of the fact that the inverse transformation of the odd Weibull family does not change its density function. Adequacy of the model for the given uncensored data is illustrated by using the plot of scaled fitted total time on test (TTT) transforms. Furthermore, simulation studies are conducted to measure the discrepancy between empirical and fitted TTT transforms by using a previously proposed test statistic. Three different examples are, respectively, providedbasedondatafromsurvival, reliabilityandenvironmentalsciencestoillustrateincreasing, bathtub and unimodal failure rates.
{"title":"Generalization of the Weibull distribution: the odd Weibull family","authors":"Kahadawala Cooray","doi":"10.1191/1471082X06st116oa","DOIUrl":"https://doi.org/10.1191/1471082X06st116oa","url":null,"abstract":"A three-parameter generalization of the Weibull distribution is presented to deal with general situations in modeling survival process with various shapes in the hazard function. This generalized Weibull distribution will be referred to as the odd Weibull family, as it is derived by considering the distributions of the odds of the Weibull and inverse Weibull families. As a result, the odd Weibull family is not only useful for testing goodness-of-fit of the Weibull and inverse Weibull as submodels, but it is also convenient for modeling and fitting different data sets, especially in the presence of censoring. The model parameters for uncensored data are estimated in two different ways because of the fact that the inverse transformation of the odd Weibull family does not change its density function. Adequacy of the model for the given uncensored data is illustrated by using the plot of scaled fitted total time on test (TTT) transforms. Furthermore, simulation studies are conducted to measure the discrepancy between empirical and fitted TTT transforms by using a previously proposed test statistic. Three different examples are, respectively, providedbasedondatafromsurvival, reliabilityandenvironmentalsciencestoillustrateincreasing, bathtub and unimodal failure rates.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130797377","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 : 2006-10-01DOI: 10.1191/1471082X06st122oa
Robert A Rigby, D. Stasinopoulos
The Box-Cox t (BCT) distribution is presented as a model for a dependent variable Y exhibiting both skewness and leptokurtosis. The distribution is defined by a power transformation Y v having a shifted and scaled (truncated) t distribution with degrees of freedom parameter τ. The distribution has four parameters and is denoted by BCT(μ, σ,ν, τ). The parameters μ, σ,ν and τ may be interpreted as relating to location (median), scale (centile-based coefficient of variation), skewness (power transformation to symmetry) and kurtosis (degrees of freedom), respectively. The generalized additive model for location, scale and shape (GAMLSS) is extended to allow each of the parameters of the distribution to be modelled as linear and/or non-linear parametric and/or smooth non-parametric functions of explanatory variables. A Fisher scoring algorithm is used to fit the model by maximizing a (penalized) likelihood. The first and expected second and cross derivatives of the likelihood with respect to μ, σ,ν and τ, required for the algorithm, are provided. The use of the BCT distribution is illustrated by two data applications.
Box-Cox t (BCT)分布是因变量Y的模型，同时显示偏度和细峰态。该分布由一个功率变换Y v定义，它具有一个移位和缩放(截断)的t分布，自由度参数为τ。分布有4个参数，用BCT(μ， σ，ν， τ)表示。参数μ、σ、ν和τ可以分别解释为与位置(中位数)、尺度(基于百分位的变异系数)、偏度(向对称的幂变换)和峰度(自由度)有关。对位置、尺度和形状的广义加性模型(GAMLSS)进行了扩展，允许将分布的每个参数建模为解释变量的线性和/或非线性参数和/或光滑非参数函数。使用Fisher评分算法通过最大化(惩罚)似然来拟合模型。给出了算法所需的似然函数对μ、σ、ν和τ的一阶导数和期望二阶导数和交叉导数。通过两个数据应用程序说明了BCT分布的使用。
{"title":"Using the Box-Cox t distribution in GAMLSS to model skewness and kurtosis","authors":"Robert A Rigby, D. Stasinopoulos","doi":"10.1191/1471082X06st122oa","DOIUrl":"https://doi.org/10.1191/1471082X06st122oa","url":null,"abstract":"The Box-Cox t (BCT) distribution is presented as a model for a dependent variable Y exhibiting both skewness and leptokurtosis. The distribution is defined by a power transformation Y v having a shifted and scaled (truncated) t distribution with degrees of freedom parameter τ. The distribution has four parameters and is denoted by BCT(μ, σ,ν, τ). The parameters μ, σ,ν and τ may be interpreted as relating to location (median), scale (centile-based coefficient of variation), skewness (power transformation to symmetry) and kurtosis (degrees of freedom), respectively. The generalized additive model for location, scale and shape (GAMLSS) is extended to allow each of the parameters of the distribution to be modelled as linear and/or non-linear parametric and/or smooth non-parametric functions of explanatory variables. A Fisher scoring algorithm is used to fit the model by maximizing a (penalized) likelihood. The first and expected second and cross derivatives of the likelihood with respect to μ, σ,ν and τ, required for the algorithm, are provided. The use of the BCT distribution is illustrated by two data applications.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116763691","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 : 2006-10-01DOI: 10.1191/1471082X06st118oa
K. Bollaerts, P. Eilers, M. Aerts
Quantile regression is an alternative to OLS regression. In quantile regression, the sum of absolute deviations or the L1-norm is minimized, whereas the sum of squared deviations or the L2-norm is minimized in OLS regression. Quantile regression has the advantage over OLS-regression of being more robust to outlying observations. Furthermore, quantile regression provides information complementing the information provided by OLS-regression. In this study, a non-parametric approach to quantile regression is presented, which constrains the estimated-quantile function to be monotone increasing. In particular, P-splines with an additional asymmetric penalty enforcing monotonicity are used within an L1-framework. This can be translated into a linear programming problem, which will be solved using an interior point algorithm. As an illustration, the presented approach will be applied to estimate quantile growth curves and quantile antibody levels as a function of age.
{"title":"Quantile regression with monotonicity restrictions using P-splines and the L1-norm","authors":"K. Bollaerts, P. Eilers, M. Aerts","doi":"10.1191/1471082X06st118oa","DOIUrl":"https://doi.org/10.1191/1471082X06st118oa","url":null,"abstract":"Quantile regression is an alternative to OLS regression. In quantile regression, the sum of absolute deviations or the L1-norm is minimized, whereas the sum of squared deviations or the L2-norm is minimized in OLS regression. Quantile regression has the advantage over OLS-regression of being more robust to outlying observations. Furthermore, quantile regression provides information complementing the information provided by OLS-regression. In this study, a non-parametric approach to quantile regression is presented, which constrains the estimated-quantile function to be monotone increasing. In particular, P-splines with an additional asymmetric penalty enforcing monotonicity are used within an L1-framework. This can be translated into a linear programming problem, which will be solved using an interior point algorithm. As an illustration, the presented approach will be applied to estimate quantile growth curves and quantile antibody levels as a function of age.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129510353","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 : 2006-10-01DOI: 10.1191/1471082X06st119oa
B. Kato, H. Hoijtink
Constrained parameter problems arise in a wide variety of applications. This article deals with estimation and model selection in linear mixed models with inequality constraints on the parameters. It is shown that different theories can be translated into statistical models by putting constraints on the model parameters yielding a set of competing models. A new approach based on the principle of encompassing priors is proposed and used to compute Bayes factors and subsequently posterior model probabilities. Model selection is based on posterior model probabilities. The approach is illustrated using a longitudinal data set.
{"title":"A Bayesian approach to inequality constrained linear mixed models: estimation and model selection","authors":"B. Kato, H. Hoijtink","doi":"10.1191/1471082X06st119oa","DOIUrl":"https://doi.org/10.1191/1471082X06st119oa","url":null,"abstract":"Constrained parameter problems arise in a wide variety of applications. This article deals with estimation and model selection in linear mixed models with inequality constraints on the parameters. It is shown that different theories can be translated into statistical models by putting constraints on the model parameters yielding a set of competing models. A new approach based on the principle of encompassing priors is proposed and used to compute Bayes factors and subsequently posterior model probabilities. Model selection is based on posterior model probabilities. The approach is illustrated using a longitudinal data set.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"205 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123253507","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 : 2006-10-01DOI: 10.1191/1471082X06st121oa
I. Plewis, F. Vitaro, R. Tremblay
Cross-informant associations tend to be low for reports of children’s behaviours at one point in time. The paper extends the literature on multiple informants using data from a well-known longitudinal study of Quebec, Canada, boys to show how to estimate associations between repeated teachers′ and self-reports of aggressive behaviour. These associations, for both level and change, are derived from multilevel models for repeated measures of variables best treated as ordered categories. The ordering is represented by sets of continuation ratios, change by linear and quadratic functions of age, and the multivariate models are estimated using penalized quasi-likelihood. The analyses also incorporate a risk variable: socio-economic status (SES). The correlations between estimates of the growth parameters for the two sets of reports tend to be rather small and smaller than the cross-informant associations for levels. SES is associated with levels of aggression, more so for teacher reports than for self-reports, but not with the decline in aggression with age.
{"title":"Modelling repeated ordinal reports from multiple informants","authors":"I. Plewis, F. Vitaro, R. Tremblay","doi":"10.1191/1471082X06st121oa","DOIUrl":"https://doi.org/10.1191/1471082X06st121oa","url":null,"abstract":"Cross-informant associations tend to be low for reports of children’s behaviours at one point in time. The paper extends the literature on multiple informants using data from a well-known longitudinal study of Quebec, Canada, boys to show how to estimate associations between repeated teachers′ and self-reports of aggressive behaviour. These associations, for both level and change, are derived from multilevel models for repeated measures of variables best treated as ordered categories. The ordering is represented by sets of continuation ratios, change by linear and quadratic functions of age, and the multivariate models are estimated using penalized quasi-likelihood. The analyses also incorporate a risk variable: socio-economic status (SES). The correlations between estimates of the growth parameters for the two sets of reports tend to be rather small and smaller than the cross-informant associations for levels. SES is associated with levels of aggression, more so for teacher reports than for self-reports, but not with the decline in aggression with age.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122582078","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：