Pub Date : 2005-04-01DOI: 10.1191/1471082X05st086oa
E. Zwane, P. V. D. van der Heijden
In the presence of continuous covariates, standard capture-recapture methods assume either that the registrations operate independently at the individual level or that the covariates can be stratified and log-linear models fitted, permitting the modelling of dependence between data sources. This article introduces an approach where direct dependence between registrations is modelled leaving the continuous covariates in their measurement scale. Simulations show that not accounting for possible dependence between registrations results in biased estimation of both the population size and standard error. The proposed method is applied to Dutch neural tube defect registration data.
{"title":"Population estimation using the multiple system estimator in the presence of continuous covariates","authors":"E. Zwane, P. V. D. van der Heijden","doi":"10.1191/1471082X05st086oa","DOIUrl":"https://doi.org/10.1191/1471082X05st086oa","url":null,"abstract":"In the presence of continuous covariates, standard capture-recapture methods assume either that the registrations operate independently at the individual level or that the covariates can be stratified and log-linear models fitted, permitting the modelling of dependence between data sources. This article introduces an approach where direct dependence between registrations is modelled leaving the continuous covariates in their measurement scale. Simulations show that not accounting for possible dependence between registrations results in biased estimation of both the population size and standard error. The proposed method is applied to Dutch neural tube defect registration data.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"1999 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128261817","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 : 2004-12-01DOI: 10.1191/1471082X04st079oa
Xeni K. Dimakos, K. Aas
In this article, we present a new approach to modelling the total economic capital required to protect a financial institution against possible losses. The approach takes into account the correlation between risk types, and in this respect, it improves upon the conventional practice that assumes perfectly correlated risks. A statistical model is built, and Monte Carlo simulation is used to estimate the total loss distribution. The methodology has been implemented in the Norwegian financial group DnB’s system for risk management. Incorporating current expert knowledge of relationships between risks, rather than taking the most conservative stand, gives a 20% reduction in the total economic capital for a one year time horizon.
{"title":"Integrated risk modelling","authors":"Xeni K. Dimakos, K. Aas","doi":"10.1191/1471082X04st079oa","DOIUrl":"https://doi.org/10.1191/1471082X04st079oa","url":null,"abstract":"In this article, we present a new approach to modelling the total economic capital required to protect a financial institution against possible losses. The approach takes into account the correlation between risk types, and in this respect, it improves upon the conventional practice that assumes perfectly correlated risks. A statistical model is built, and Monte Carlo simulation is used to estimate the total loss distribution. The methodology has been implemented in the Norwegian financial group DnB’s system for risk management. Incorporating current expert knowledge of relationships between risks, rather than taking the most conservative stand, gives a 20% reduction in the total economic capital for a one year time horizon.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115348120","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 : 2004-12-01DOI: 10.1191/1471082X04st082oa
C. Pfeifer
A semi-parametric model is applied in order to model counts of letters for the federal Austrian postal system. Random coefficients are introduced into the splined variable of the semi-parametric regression model to describe heterogeneity of the temporal effect. Pfeifer and Seeber propose estimates for random coefficients to classify post offices by a hierarchical cluster algorithm. In this article, we apply two model based approaches for classification. It turns out here that both the hierarchical and the model based approach are useful for explorative cluster analysis.
{"title":"Classification of longitudinal profiles based on semi-parametric regression with mixed effects","authors":"C. Pfeifer","doi":"10.1191/1471082X04st082oa","DOIUrl":"https://doi.org/10.1191/1471082X04st082oa","url":null,"abstract":"A semi-parametric model is applied in order to model counts of letters for the federal Austrian postal system. Random coefficients are introduced into the splined variable of the semi-parametric regression model to describe heterogeneity of the temporal effect. Pfeifer and Seeber propose estimates for random coefficients to classify post offices by a hierarchical cluster algorithm. In this article, we apply two model based approaches for classification. It turns out here that both the hierarchical and the model based approach are useful for explorative cluster analysis.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121297156","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 : 2004-12-01DOI: 10.1191/1471082X04st083oa
J. Singer, J. Nobre, Henry Corazza Sef
We consider regression models with no intercepts to analyse pretest/posttest data from a dental study conducted under an experimental design involving a blocked factorial structure with two within individual factors. The proposed models accommodate block effects, heteroscedasticity, nonlinear relations between pretest and posttest measures and repeated measures. We compare multiplicative lognormal and gamma models to additive normal models fitted via generalized linear models methodology for repeated measures. Alternatively, we consider standard linear mixed models methodology to fit lognormal models, an option that facilitates modelling the within subjects covariance structure.
{"title":"Regression models for pretest/posttest data in blocks","authors":"J. Singer, J. Nobre, Henry Corazza Sef","doi":"10.1191/1471082X04st083oa","DOIUrl":"https://doi.org/10.1191/1471082X04st083oa","url":null,"abstract":"We consider regression models with no intercepts to analyse pretest/posttest data from a dental study conducted under an experimental design involving a blocked factorial structure with two within individual factors. The proposed models accommodate block effects, heteroscedasticity, nonlinear relations between pretest and posttest measures and repeated measures. We compare multiplicative lognormal and gamma models to additive normal models fitted via generalized linear models methodology for repeated measures. Alternatively, we consider standard linear mixed models methodology to fit lognormal models, an option that facilitates modelling the within subjects covariance structure.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123221517","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 : 2004-12-01DOI: 10.1191/1471082X04st081oa
J. Little, M. Goldstein, P. Jonathan
Efficient inspection and maintenance of complex industrial systems, subject to degradation effects such as corrosion, are important for safety and economic reasons. With appropriate statistical modelling, the utilization of inspection resources and the quality of inferences can be greatly improved. We develop a suitable Bayesian spatio-temporal dynamic linear model for problems such as wall thickness monitoring. We are concerned with problems where the inspection method used collects transformed data, for example minimum regional remaining wall thicknesses. We describe how the model may be used to derive efficient inspection schedules by identifying when, where and how much inspection should be made in the future.
{"title":"Efficient Bayesian sampling inspection for industrial processes based on transformed spatio-temporal data","authors":"J. Little, M. Goldstein, P. Jonathan","doi":"10.1191/1471082X04st081oa","DOIUrl":"https://doi.org/10.1191/1471082X04st081oa","url":null,"abstract":"Efficient inspection and maintenance of complex industrial systems, subject to degradation effects such as corrosion, are important for safety and economic reasons. With appropriate statistical modelling, the utilization of inspection resources and the quality of inferences can be greatly improved. We develop a suitable Bayesian spatio-temporal dynamic linear model for problems such as wall thickness monitoring. We are concerned with problems where the inspection method used collects transformed data, for example minimum regional remaining wall thicknesses. We describe how the model may be used to derive efficient inspection schedules by identifying when, where and how much inspection should be made in the future.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123639991","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 : 2004-12-01DOI: 10.1191/1471082X04st080oa
I. Currie, M. Durbán, P. Eilers
The prediction of future mortality rates is a problem of fundamental importance for the insurance and pensions industry. We show how the method of P-splines can be extended to the smoothing and forecasting of two-dimensional mortality tables. We use a penalized generalized linear model with Poisson errors and show how to construct regression and penalty matrices appropriate for two-dimensional modelling. An important feature of our method is that forecasting is a natural consequence of the smoothing process. We illustrate our methods with two data sets provided by the Continuous Mortality Investigation Bureau, a central body for the collection and processing of UK insurance and pensions data.
{"title":"Smoothing and forecasting mortality rates","authors":"I. Currie, M. Durbán, P. Eilers","doi":"10.1191/1471082X04st080oa","DOIUrl":"https://doi.org/10.1191/1471082X04st080oa","url":null,"abstract":"The prediction of future mortality rates is a problem of fundamental importance for the insurance and pensions industry. We show how the method of P-splines can be extended to the smoothing and forecasting of two-dimensional mortality tables. We use a penalized generalized linear model with Poisson errors and show how to construct regression and penalty matrices appropriate for two-dimensional modelling. An important feature of our method is that forecasting is a natural consequence of the smoothing process. We illustrate our methods with two data sets provided by the Continuous Mortality Investigation Bureau, a central body for the collection and processing of UK insurance and pensions data.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116524341","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 : 2004-12-01DOI: 10.1191/1471082X04st077ed
A. Frigessi, J. Engel
The ENBIS organization (European Network for Business and Industrial Statistics, www.enbis.org) is a network for statistical practitioners in European business and industrial environments and it aims to promote the widespread use of applied statistical methods. Statistics in business and industry implies the use of a wide range of quantitative methods. Statistical modelling is often a basic tool to get insight into the running process, and to find the main characteristics of a designed product. The aim of this special ENBIS issue of Statistical Modelling: an International Journal is to show new results of the use of statistical modelling in business and industry, to promote the development of these models and to stimulate practitioners to find new applications for them.
{"title":"Editorial: a special ENBIS issue of SMIJ","authors":"A. Frigessi, J. Engel","doi":"10.1191/1471082X04st077ed","DOIUrl":"https://doi.org/10.1191/1471082X04st077ed","url":null,"abstract":"The ENBIS organization (European Network for Business and Industrial Statistics, www.enbis.org) is a network for statistical practitioners in European business and industrial environments and it aims to promote the widespread use of applied statistical methods. Statistics in business and industry implies the use of a wide range of quantitative methods. Statistical modelling is often a basic tool to get insight into the running process, and to find the main characteristics of a designed product. The aim of this special ENBIS issue of Statistical Modelling: an International Journal is to show new results of the use of statistical modelling in business and industry, to promote the development of these models and to stimulate practitioners to find new applications for them.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125461572","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 : 2004-12-01DOI: 10.1191/1471082X04st078oa
S. Kuhnt, M. Erdbrügge
In this article, we provide a strategy for the simultaneous optimization of multiple responses. Cases are covered where a set of response variables has finite target values and depends on easy to control as well as on hard to control variables. Our approach is based on loss functions, without the need for a predefined cost matrix. For each element of a sequence of possible weights assigned to the individual responses, settings of the easy to control parameters are determined, which minimize the estimated mean of a multivariate loss function. The estimation is based on statistical models, which depend only on the easy to control variables. The loss function itself takes the value zero, if all responses are on target with zero variances. In each case, the derived parameter settings are connected to a specific compromise of the responses, which is graphically displayed to the engineer by so called joint optimization plots. The expert can thereby gain valuable insight into the production process and then decide on the most sensible parameter setting. The proposed strategy is illustrated with a data set from the literature and new data from an up to date application.
{"title":"A strategy of robust parameter design for multiple responses","authors":"S. Kuhnt, M. Erdbrügge","doi":"10.1191/1471082X04st078oa","DOIUrl":"https://doi.org/10.1191/1471082X04st078oa","url":null,"abstract":"In this article, we provide a strategy for the simultaneous optimization of multiple responses. Cases are covered where a set of response variables has finite target values and depends on easy to control as well as on hard to control variables. Our approach is based on loss functions, without the need for a predefined cost matrix. For each element of a sequence of possible weights assigned to the individual responses, settings of the easy to control parameters are determined, which minimize the estimated mean of a multivariate loss function. The estimation is based on statistical models, which depend only on the easy to control variables. The loss function itself takes the value zero, if all responses are on target with zero variances. In each case, the derived parameter settings are connected to a specific compromise of the responses, which is graphically displayed to the engineer by so called joint optimization plots. The expert can thereby gain valuable insight into the production process and then decide on the most sensible parameter setting. The proposed strategy is illustrated with a data set from the literature and new data from an up to date application.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121668410","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 : 2004-10-01DOI: 10.1191/1471082X04st073oa
A. Gannoun, J. Saracco, W. Urfer, G. Bonney
Microarrays are part of a new class of biotechnologies, which allow the monitoring of expression levels of thousands of genes simultaneously. In microarray data analysis, the comparison of gene expression profiles with respect to different conditions and the selection of biologically interesting genes are crucial tasks. Multivariate statistical methods have been applied to analyze these large data sets. To identify genes with altered expression under two experimental conditions, we propose a nonparametric statistical approach. Specifically, we propose estimating the distributions of a t-type statistic and its null statistic, using kernel methods. A comparison of these two distributions by means of a likelihood ratio test can identify genes with significantly changed expressions. A new method to provide more stable estimates of tail probabilities is proposed, as well as a method for the calculation of the cut-off point and the acceptance region. The methodology is applied to a leukaemia data set containing expression levels of 7129 genes, and is compared with normal mixture model and the traditional t-test.
{"title":"Nonparametric analysis of replicated microarray experiments","authors":"A. Gannoun, J. Saracco, W. Urfer, G. Bonney","doi":"10.1191/1471082X04st073oa","DOIUrl":"https://doi.org/10.1191/1471082X04st073oa","url":null,"abstract":"Microarrays are part of a new class of biotechnologies, which allow the monitoring of expression levels of thousands of genes simultaneously. In microarray data analysis, the comparison of gene expression profiles with respect to different conditions and the selection of biologically interesting genes are crucial tasks. Multivariate statistical methods have been applied to analyze these large data sets. To identify genes with altered expression under two experimental conditions, we propose a nonparametric statistical approach. Specifically, we propose estimating the distributions of a t-type statistic and its null statistic, using kernel methods. A comparison of these two distributions by means of a likelihood ratio test can identify genes with significantly changed expressions. A new method to provide more stable estimates of tail probabilities is proposed, as well as a method for the calculation of the cut-off point and the acceptance region. The methodology is applied to a leukaemia data set containing expression levels of 7129 genes, and is compared with normal mixture model and the traditional t-test.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127889571","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 : 2004-10-01DOI: 10.1191/1471082X04st072oa
R. Dittrich, R. Hatzinger, W. Katzenbeisser
The purpose of this paper is to propose an alternative log-linear representation of an adjacent categories (AC) paired comparison (PC) model. The AC model is well suited for modelling ordinal PC data by postulating a power relationship between the response category and the probability of preferring one object over another object. The model is applied to data collected on the motivation of Vienna students to start a doctoral programme of study.
{"title":"A log-linear approach for modelling ordinal paired comparison data on motives to start a PhD programme","authors":"R. Dittrich, R. Hatzinger, W. Katzenbeisser","doi":"10.1191/1471082X04st072oa","DOIUrl":"https://doi.org/10.1191/1471082X04st072oa","url":null,"abstract":"The purpose of this paper is to propose an alternative log-linear representation of an adjacent categories (AC) paired comparison (PC) model. The AC model is well suited for modelling ordinal PC data by postulating a power relationship between the response category and the probability of preferring one object over another object. The model is applied to data collected on the motivation of Vienna students to start a doctoral programme of study.","PeriodicalId":354759,"journal":{"name":"Statistical Modeling","volume":"208 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131618362","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：