Loading

Error: Cannot Load Popup Box

Hit List

Title:

Modeling outliers, bursts and flat stretches in time series using Mixture Transition Distribution (MTD) models

Author:

Description:

The class of Mixture Transition Distribution (MTD) time series models is introduced. In these models, the conditional distribution of the current observation given the past is a mixture of conditional distributions given each one of the last p observations. They can capture non-Gaussian and non-linear features such as outliers, bursts of activit...

The class of Mixture Transition Distribution (MTD) time series models is introduced. In these models, the conditional distribution of the current observation given the past is a mixture of conditional distributions given each one of the last p observations. They can capture non-Gaussian and non-linear features such as outliers, bursts of activity and flat stretches, in a single unified model class. They can also represent time series defined on arbitrary state spaces, which need not even be Euclidean. They perform well in the usual case of Gaussian time series without obvious non-standard behaviors. The models are simple, analytically tractable, easy to simulate and readily estimated. The stationarity and autocorrelation properties of the models are derived. _A. simple EM algorithm is given and shown to work well for estimation. The models are applied to several real and simulated data sets with satisfactory results. They appear to capture the features of the data better than the best competing ARIMA models. Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2009-08-06

Source:

http://www.stat.washington.edu/research/reports/1990/tr194.pdf

http://www.stat.washington.edu/research/reports/1990/tr194.pdf Minimize

Document Type:

text

Language:

en

DDC:

519 Probabilities & applied mathematics *(computed)*

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

Metadata may be used without restrictions as long as the oai identifier remains attached to it. Minimize

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

References Open Access Rapid responses

Author:

Description:

Bayesian melding for estimating uncertainty in

Bayesian melding for estimating uncertainty in Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2013-07-17

Source:

http://www.stat.washington.edu/people/raftery/Research/PDF/AlkemaEtal2008STI.pdf

http://www.stat.washington.edu/people/raftery/Research/PDF/AlkemaEtal2008STI.pdf Minimize

Document Type:

text

Language:

en

Subjects:

Statistics and the Social

Statistics and the Social Minimize

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

Metadata may be used without restrictions as long as the oai identifier remains attached to it. Minimize

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

How many clusters? Which clustering method? Answers via model-based cluster analysis

Author:

Description:

We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian comp...

We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian components with different parameterizations and cross-cluster constraints. Noise and outliers can be modeled by adding a Poisson process component. Partitions are determined by the EM (expectation-maximization) algorithm for maximum likelihood, with initial values from agglomerative hierarchical clustering. Models are compared using an approximation to the Bayes factor based on the Bayesian Information Criterion (BIC); unlike significance tests, this allows comparison of more than two models at the same time, and removes the restriction that the models compared be nested. The problems of determining the number of clusters and the clustering method are solved simultaneously by choosing the best model. Moreover, the EM result provides a measure of uncertainty about the associated classification of each data point. Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2011-10-29

Source:

http://www.albany.edu/~jz7088/documents/03-18-2004/fraley.clustrng.no.pdf

http://www.albany.edu/~jz7088/documents/03-18-2004/fraley.clustrng.no.pdf Minimize

Document Type:

text

Language:

en

DDC:

310 Collections of general statistics *(computed)*

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

Metadata may be used without restrictions as long as the oai identifier remains attached to it. Minimize

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

Choosing the Link Function and Accounting for Link Uncertainty in Generalized Linear Models using Bayes Factors

Author:

Description:

this paper, we extend the approach taken by Raftery (1996) to calculate approximate Bayes factors for GLM's with a parametric link function. Even though GLM's with canonical links (for de nition see McCullagh and Nelder (1989)), such as the logit link in binomial regression, guarantee maximum information and a simple interpretation of the regres...

this paper, we extend the approach taken by Raftery (1996) to calculate approximate Bayes factors for GLM's with a parametric link function. Even though GLM's with canonical links (for de nition see McCullagh and Nelder (1989)), such as the logit link in binomial regression, guarantee maximum information and a simple interpretation of the regression parameters, they do not always provide the best t available to a given data set. Link misspeci - cation can lead to substantial bias in the regression parameters and the mean response estimates (see Czado and Santner (1992) for binomial responses). One common approach to guard against link misspeci cation in generalized linear models is to embed the canonical link in a wide parametric class of links = = fF (; ); 2 g, which includes the canonical link as a special case when = 0 . Many such parametric link classes for binary regression data have been proposed in the literature. Montfort and Otten (1976), Copenhaver and Mielke (1977), Aranda-Ordaz (1981) , Guerrero and Johnson (1982), Morgan (1983) and Whittmore (1983) proposed one-parameter families, while Prentice (1976), Pregibon (1980), Stukel (1988) and Czado (1992) considered two-parameter families. Link functions for the non-binary case were studied by Pregibon (1980) and Czado (1992, 1997) Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2009-04-17

Source:

http://www-m4.mathematik.tu-muenchen.de/m4/Papers/Czado/bayes.ps

http://www-m4.mathematik.tu-muenchen.de/m4/Papers/Czado/bayes.ps Minimize

Document Type:

text

Language:

en

Subjects:

Key words ; Bayes factors ; link function ; GLM ; model selection ; reference

Key words ; Bayes factors ; link function ; GLM ; model selection ; reference Minimize

DDC:

310 Collections of general statistics *(computed)*

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

How many clusters? Which clustering method? Answers via model-based cluster analysis

Author:

Description:

0330. Thanks go to Simon Byers for providing the NNclean denoising procedure. We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluste...

0330. Thanks go to Simon Byers for providing the NNclean denoising procedure. We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian components with different parameterizations and cross-cluster constraints. Noise and outliers can be modeled by adding a Poisson process component. Partitions are determined by the EM (expectation-maximization) algorithm for maximum likelihood, with initial values from agglomerative hierarchical clustering. Models are compared using an approximation to the Bayes factor based on the Bayesian Information Criterion (BIC); unlike significance tests, this allows comparison of more than two models at the same time, and removes the restriction that the models compared be nested. The problems of determining the number of clusters and the clustering method are solved simultaneously by choosing the best model. Moreover, the EM result provides a Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2008-07-17

Source:

http://www.ics.uci.edu/~smyth/courses/ics274/fraley_raftery.pdf

http://www.ics.uci.edu/~smyth/courses/ics274/fraley_raftery.pdf Minimize

Document Type:

text

Language:

en

Subjects:

Contents

Contents Minimize

DDC:

310 Collections of general statistics *(computed)*

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

Demography DOI 10.1007/s13524-012-0193-x Bayesian Probabilistic Projections of Life Expectancy for All Countries

Author:

Description:

© The Author(s) 2013. This article is published with open access at Springerlink.com Abstract We propose a Bayesian hierarchical model for producing probabilistic forecasts of male period life expectancy at birth for all the countries of the world to 2100. Such forecasts would be an input to the production of probabilistic population projections...

© The Author(s) 2013. This article is published with open access at Springerlink.com Abstract We propose a Bayesian hierarchical model for producing probabilistic forecasts of male period life expectancy at birth for all the countries of the world to 2100. Such forecasts would be an input to the production of probabilistic population projections for all countries, which is currently being considered by the United Nations. To evaluate the method, we conducted an out-of-sample cross-validation experiment, fitting the model to the data from 1950–1995 and using the estimated model to forecast for the subsequent 10 years. The 10-year predictions had a mean absolute error of about 1 year, about 40 % less than the current UN methodology. The probabilistic forecasts were calibrated in the sense that, for example, the 80 % prediction intervals contained the truth about 80 % of the time. We illustrate our method with results from Madagascar (a typical country with steadily improving life expectancy), Latvia (a country that has had a mortality crisis), and Japan (a leading country). We also show aggregated results for South Asia, a region with eight countries. Free, publicly available R software packages called bayesLife and bayesDem are available to implement the method. Electronic supplementary material The online version of this article (doi: 10.1007/s13524-012-0193-x) contains supplementary material, which is available to authorized users. Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2013-07-15

Source:

http://www.stat.washington.edu/raftery/Research/PDF/RafteryChunn2013.pdf

Document Type:

text

Language:

en

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

How many clusters? Which clustering method? Answers via model-based cluster analysis

Author:

Description:

0330. Thanks go to Simon Byers for providing the NNclean denoising procedure. We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluste...

0330. Thanks go to Simon Byers for providing the NNclean denoising procedure. We consider the problem of determining the structure of clustered data, without prior knowledge of the number of clusters or any other information about their composition. Data are represented by a mixture model in which each component corresponds to a different cluster. Models with varying geometric properties are obtained through Gaussian components with different parameterizations and cross-cluster constraints. Noise and outliers can be modeled by adding a Poisson process component. Partitions are determined by the EM (expectation-maximization) algorithm for maximum likelihood, with initial values from agglomerative hierarchical clustering. Models are compared using an approximation to the Bayes factor based on the Bayesian Information Criterion (BIC); unlike significance tests, this allows comparison of more than two models at the same time, and removes the restriction that the models compared be nested. The problems of determining the number of clusters and the clustering method are solved simultaneously by choosing the best model. Moreover, the EM result provides a Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2009-01-06

Source:

http://www.ics.uci.edu/~smyth/courses/ics278/papers/fraley_clustering.pdf

http://www.ics.uci.edu/~smyth/courses/ics278/papers/fraley_clustering.pdf Minimize

Document Type:

text

Language:

en

Subjects:

Contents

Contents Minimize

DDC:

310 Collections of general statistics *(computed)*

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

Correspondence to:

Author:

Description:

Bayesian melding for estimating uncertainty in

Bayesian melding for estimating uncertainty in Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2013-04-04

Source:

ftp://ftp.ncbi.nlm.nih.gov/pub/pmc/86/65/Sex_Transm_Infect_2008_Aug_22_84(Suppl_1)_i11-i16.tar.gz

ftp://ftp.ncbi.nlm.nih.gov/pub/pmc/86/65/Sex_Transm_Infect_2008_Aug_22_84(Suppl_1)_i11-i16.tar.gz Minimize

Document Type:

text

Language:

en

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

MCLUST: Software for Model-Based Cluster Analysis

Author:

Description:

ents the data, and k is an integer subscript specifying a particular cluster. Clusters are ellispoidal, centered at the means ¯ k . The covariances \Sigma k determine their other geometric features. Each covariance matrix is parameterized by eigenvalue decomposition in the form \Sigma k = k D k A k D T k ; Funded by the Office of Naval Research ...

ents the data, and k is an integer subscript specifying a particular cluster. Clusters are ellispoidal, centered at the means ¯ k . The covariances \Sigma k determine their other geometric features. Each covariance matrix is parameterized by eigenvalue decomposition in the form \Sigma k = k D k A k D T k ; Funded by the Office of Naval Research under contracts N00014-96-1-0192 and N00014-96-1-0330. 1 MathSoft, Inc., Seattle, WA USA --- http://www.mathsoft.com/splus where D k is the orthogonal matrix of eigenvectors, A k is a diagonal matrix whose elements are proportional to the eigenvalues of \Sigma k , and k is a scalar. The orientation of the principal components of \Sigma k is deter Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2009-04-11

Source:

http://www.stat.unipg.it/stat/statlib/S/mclust/mclust.ps.gz

http://www.stat.unipg.it/stat/statlib/S/mclust/mclust.ps.gz Minimize

Document Type:

text

Language:

en

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Title:

MCLUST: Software for Model-Based Cluster Analysis

Author:

Description:

ents the data, and k is an integer subscript specifying a particular cluster. Clusters are ellipsoidal, centered at the means ¯ k . The covariances \Sigma k determine their other geometric features. Each covariance matrix is parameterized by eigenvalue decomposition in the form \Sigma k = k D k A k D T k ; Funded by the Office of Naval Research ...

ents the data, and k is an integer subscript specifying a particular cluster. Clusters are ellipsoidal, centered at the means ¯ k . The covariances \Sigma k determine their other geometric features. Each covariance matrix is parameterized by eigenvalue decomposition in the form \Sigma k = k D k A k D T k ; Funded by the Office of Naval Research under contracts N00014-96-1-0192 and N00014-96-1-0330. 1 MathSoft, Inc., Seattle, WA USA --- http://www.mathsoft.com/splus where D k is the orthogonal matrix of eigenvectors, A k is a diagonal matrix whose elements are proportional to the eigenvalues of \Sigma k , and k is a scalar. The orientation of the principal components of \Sigma k is deter Minimize

Contributors:

The Pennsylvania State University CiteSeerX Archives

Year of Publication:

2009-04-11

Source:

http://www.stat.washington.edu/tech.reports/tr342.ps

http://www.stat.washington.edu/tech.reports/tr342.ps Minimize

Document Type:

text

Language:

en

Rights:

Metadata may be used without restrictions as long as the oai identifier remains attached to it.

URL:

Content Provider:

My Lists:

My Tags:

Notes:

Currently in BASE: 68,072,316 Documents of 3,307 Content Sources

http://www.base-search.net