Open Access. Powered by Scholars. Published by Universities.®

Statistical Theory Commons™

Open Access. Powered by Scholars. Published by Universities.®

Discipline
Institution
Keyword
Publication Year
Publication
Publication Type

Articles 1171 - 1200 of 1633

Full-Text Articles in Statistical Theory

Extending Marginal Structural Models Through Local, Penalized, And Additive Learning, Daniel Rubin, Mark J. Van Der Laan Sep 2006

Extending Marginal Structural Models Through Local, Penalized, And Additive Learning, Daniel Rubin, Mark J. Van Der Laan

U.C. Berkeley Division of Biostatistics Working Paper Series

Marginal structural models (MSMs) allow one to form causal inferences from data, by specifying a relationship between a treatment and the marginal distribution of a corresponding counterfactual outcome. Following their introduction in Robins (1997), MSMs have typically been fit after assuming a semiparametric model, and then estimating a finite dimensional parameter. van der Laan and Dudoit (2003) proposed to instead view MSM fitting not as a task of semiparametric parameter estimation, but of nonparametric function approximation. They introduced a class of causal effect estimators based on mapping loss functions suitable for the unavailable counterfactual data to those suitable for the …

Go to article

Statistical Learning Of Origin-Specific Statically Optimal Individualized Treatment Rules, Mark J. Van Der Laan, Maya L. Petersen Sep 2006

Statistical Learning Of Origin-Specific Statically Optimal Individualized Treatment Rules, Mark J. Van Der Laan, Maya L. Petersen

U.C. Berkeley Division of Biostatistics Working Paper Series

Consider a longitudinal observational or controlled study in which one collects chronological data over time on n randomly sampled subjects. The time-dependent process one observes on each randomly sampled subject contains time-dependent covariates, time-dependent treatment actions, and an outcome process or single final outcome of interest. A statically optimal individualized treatment rule (as introduced in van der Laan, Petersen & Joffe (2005), Petersen & van der Laan (2006)) is a (unknown) treatment rule which at any point in time conditions on a user-supplied subset of the past, computes the future static treatment regimen that maximizes a (conditional) mean future outcome …

Go to article

Comparing The Statistical Tests For Homogeneity Of Variances., Zhiqiang Mu Aug 2006

Comparing The Statistical Tests For Homogeneity Of Variances., Zhiqiang Mu

Electronic Theses and Dissertations

Testing the homogeneity of variances is an important problem in many applications since statistical methods of frequent use, such as ANOVA, assume equal variances for two or more groups of data. However, testing the equality of variances is a difficult problem due to the fact that many of the tests are not robust against non-normality. It is known that the kurtosis of the distribution of the source data can affect the performance of the tests for variance. We review the classical tests and their latest, more robust modifications, some other tests that have recently appeared in the literature, and use …

Go to article

Predicting Future Responses Based On Possibly Misspecified Working Models, Tianxi Cai, Lu Tian, Scott D. Solomon, L.J. Wei Aug 2006

Predicting Future Responses Based On Possibly Misspecified Working Models, Tianxi Cai, Lu Tian, Scott D. Solomon, L.J. Wei

Harvard University Biostatistics Working Paper Series

No abstract provided.

Go to article

The Combination Of Ecological And Case-Control Data, Sebastien Haneuse, Jon Wakefield Jul 2006

The Combination Of Ecological And Case-Control Data, Sebastien Haneuse, Jon Wakefield

UW Biostatistics Working Paper Series

Ecological studies, in which data are available at the level of the group, rather than at the level of the individual, are susceptible to a range of biases due to their inability to characterize within-group variability in exposures and confounders. In order to overcome these biases, we propose a hybrid design in which ecological data are supplemented with a sample of individual-level case-control data. We develop the likelihood for this design and illustrate its benefits via simulation, both in bias reduction when compared to an ecological study, and in efficiency gains relative to a conventional case-control study. An interesting special …

Go to article

The Combination Of Ecological And Case-Control Data, Sebastien Haneuse, Jon Wakefield Jul 2006

The Combination Of Ecological And Case-Control Data, Sebastien Haneuse, Jon Wakefield

UW Biostatistics Working Paper Series

Ecological studies, in which data are available at the level of the group, rather than at the level of the individual, are susceptible to a range of biases due to their inability to characterize within-group variability in exposures and confounders. In order to overcome these biases, we propose a hybrid design in which ecological data are supplemented with a sample of individual-level case-control data. We develop the likelihood for this design and illustrate its benefits via simulation, both in bias reduction when compared to an ecological study, and in efficiency gains relative to a conventional case-control study. An interesting special …

Go to article

Doubly Robust Censoring Unbiased Transformations, Daniel Rubin, Mark J. Van Der Laan Jun 2006

Doubly Robust Censoring Unbiased Transformations, Daniel Rubin, Mark J. Van Der Laan

U.C. Berkeley Division of Biostatistics Working Paper Series

We consider random design nonparametric regression when the response variable is subject to right censoring. Following the work of Fan and Gijbels (1994), a common approach to this problem is to apply what has been termed a censoring unbiased transformation to the data to obtain surrogate responses, and then enter these surrogate responses with covariate data into standard smoothing algorithms. Existing censoring unbiased transformations generally depend on either the conditional survival function of the response of interest, or that of the censoring variable. We show that a mapping introduced in another statistical context is in fact a censoring unbiased transformation …

Go to article

A Method To Increase The Power Of Multiple Testing Procedures Through Sample Splitting, Daniel Rubin, Sandrine Dudoit, Mark J. Van Der Laan Jun 2006

A Method To Increase The Power Of Multiple Testing Procedures Through Sample Splitting, Daniel Rubin, Sandrine Dudoit, Mark J. Van Der Laan

U.C. Berkeley Division of Biostatistics Working Paper Series

Consider the standard multiple testing problem where many hypotheses are to be tested, each hypothesis is associated with a test statistic, and large test statistics provide evidence against the null hypotheses. One proposal to provide probabilistic control of Type-I errors is the use of procedures ensuring that the expected number of false positives does not exceed a user-supplied threshold. Among such multiple testing procedures, we derive the ``most powerful'' method, meaning the test statistic cutoffs that maximize the expected number of true positives. Unfortunately, these optimal cutoffs depend on the true unknown data generating distribution, so could never be used …

Go to article

Bayesian Reference Inference On The Ratio Of Poisson Rates., Changbin Guo May 2006

Bayesian Reference Inference On The Ratio Of Poisson Rates., Changbin Guo

Electronic Theses and Dissertations

Bayesian reference analysis is a method of determining the prior under the Bayesian paradigm. It incorporates as little information as possible from the experiment. Estimation of the ratio of two independent Poisson rates is a common practical problem. In this thesis, the method of reference analysis is applied to derive the posterior distribution of the ratio of two independent Poisson rates, and then to construct point and interval estimates based on the reference posterior. In addition, the Frequentist coverage property of HPD intervals is verified through simulation.

Go to article

Confidence Intervals For An Effect Size When Variances Are Not Equal, James Algina, H. J. Keselman, Randall D. Penfield May 2006

Confidence Intervals For An Effect Size When Variances Are Not Equal, James Algina, H. J. Keselman, Randall D. Penfield

Journal of Modern Applied Statistical Methods

Confidence intervals must be robust in having nominal and actual probability coverage in close agreement. This article examined two ways of computing an effect size in a two-group problem: (a) the classic approach which divides the mean difference by a single standard deviation and (b) a variant of a method which replaces least squares values with robust trimmed means and a Winsorized variance. Confidence intervals were determined with theoretical and bootstrap critical values. Only the method that used robust estimators and a bootstrap critical value provided generally accurate probability coverage under conditions of nonnormality and variance heterogeneity in balanced as …

Go to article

Limitations Of The Analysis Of Variance, Phillip I. Good, Clifford E. Lunneborg May 2006

Limitations Of The Analysis Of Variance, Phillip I. Good, Clifford E. Lunneborg

Journal of Modern Applied Statistical Methods

Conditions under which the analysis of variance will yield inexact p-values or would be inferior in power to a permutation test are investigated. The findings for the one-way design are consistent with and extend those of Miller (1980).

Go to article

Ancova: A Robust Omnibus Test Based On Selected Design Points, Rand R. Wilcox May 2006

Ancova: A Robust Omnibus Test Based On Selected Design Points, Rand R. Wilcox

Journal of Modern Applied Statistical Methods

Many robust analogs of the classic analysis of covariance method have been proposed. One approach, when comparing two independent groups, uses selected design points and then compares the groups at each design point using some robust method for comparing measures of location. So, if K design points are of interest, K tests are performed. There are rather obvious ways of performing, instead, an omnibus test that for all K points, no differences between the groups exist. One of the main results here is that several variations of these methods can perform very poorly in simulations. An alternative approach, based in …

Go to article

Penalized Splines For Longitudinal Data With An Application In Aids Studies, Hua Liang, Yuanhui Xiao May 2006

Penalized Splines For Longitudinal Data With An Application In Aids Studies, Hua Liang, Yuanhui Xiao

Journal of Modern Applied Statistical Methods

A penalized spline approximation is proposed in considering nonparametric regression for longitudinal data. Standard linear mixed-effects modeling can be applied for the estimation. It is relatively simple, efficiently computed, and robust to the smooth parameters selection, which are often encountered when local polynomial and smoothing spline techniques are used to analyze longitudinal data set. The method is extended to time-varying coefficient mixed-effects models. The proposed methods are applied to data from an AIDS clinical study. Biological interpretations and clinical implications are discussed. Simulation studies are done to illustrate the proposed methods.

Go to article

Choosing Smoothing Parameters For Exponential Smoothing: Minimizing Sums Of Squared Versus Sums Of Absolute Errors, Terry E. Dielman May 2006

Choosing Smoothing Parameters For Exponential Smoothing: Minimizing Sums Of Squared Versus Sums Of Absolute Errors, Terry E. Dielman

Journal of Modern Applied Statistical Methods

When choosing smoothing parameters in exponential smoothing, the choice can be made by either minimizing the sum of squared one-step-ahead forecast errors or minimizing the sum of the absolute onestep- ahead forecast errors. In this article, the resulting forecast accuracy is used to compare these two options.

Go to article

The Efficiency Of Ols In The Presence Of Auto-Correlated Disturbances In Regression Models, Samir Safi, Alexander White May 2006

The Efficiency Of Ols In The Presence Of Auto-Correlated Disturbances In Regression Models, Samir Safi, Alexander White

Journal of Modern Applied Statistical Methods

The ordinary least squares (OLS) estimates in the regression model are efficient when the disturbances have mean zero, constant variance, and are uncorrelated. In problems concerning time series, it is often the case that the disturbances are correlated. Using computer simulations, the robustness of various estimators are considered, including estimated generalized least squares. It was found that if the disturbance structure is autoregressive and the dependent variable is nonstochastic and linear or quadratic, the OLS performs nearly as well as its competitors. For other forms of the dependent variable, rules of thumb are presented to guide practitioners in the choice …

Go to article

Understanding Eurasian Convergence: Application Of Kohonen Self-Organizing Maps, Joel I. Deichmann, Abdolreza Eshghi, Dominique Haughton, Selin Sayek, Nicholas Teebagy, Heikki Topi May 2006

Understanding Eurasian Convergence: Application Of Kohonen Self-Organizing Maps, Joel I. Deichmann, Abdolreza Eshghi, Dominique Haughton, Selin Sayek, Nicholas Teebagy, Heikki Topi

Journal of Modern Applied Statistical Methods

Kohonen self-organizing maps (SOMs) are employed to examine economic and social convergence of Eurasian countries based on a set of twenty-eight socio-economic measures. A core of European Union states is identified that provides a benchmark against which convergence of post-socialist transition economies may be judged. The Central European Visegrád countries and Baltics show the greatest economic convergence to Western Europe, while other states form clusters that lag behind. Initial conditions on the social dimension can either facilitate or constrain economic convergence, as discovered in Central Europe vis-à-vis the Central Asian Republics. Disquiet in the convergence literature is resolved by providing …

Go to article

Analysis Of Type-Ii Progressively Hybrid Censored Competing Risks Data, Debasis Kundu, Avijit Joarder May 2006

Analysis Of Type-Ii Progressively Hybrid Censored Competing Risks Data, Debasis Kundu, Avijit Joarder

Journal of Modern Applied Statistical Methods

A Type-II progressively hybrid censoring scheme for competing risks data is introduced, where the experiment terminates at a pre-specified time. The likelihood inference of the unknown parameters is derived under the assumptions that the lifetime distributions of the different causes are independent and exponentially distributed. The maximum likelihood estimators of the unknown parameters are obtained in exact forms. Asymptotic confidence intervals and two bootstrap confidence intervals are also proposed. Bayes estimates and credible intervals of the unknown parameters are obtained under the assumption of gamma priors on the unknown parameters. Different methods have been compared using Monte Carlo simulations. One …

Go to article

Jmasm23: Cluster Analysis In Epidemiological Data (Matlab), Andrés M. Alonso May 2006

Jmasm23: Cluster Analysis In Epidemiological Data (Matlab), Andrés M. Alonso

Journal of Modern Applied Statistical Methods

Matlab functions for testing the existence of time, space and time-space clusters of disease occurrences are presented. The classical scan test, the Ederer, Myers and Mantel’s test, the Ohno, Aoki and Aoki’s test, and the Knox’s test are considered.

Go to article

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes, Edward J. Mascha, Jeffrey M. Albert May 2006

Properties Of Bound Estimators On Treatment Effect Heterogeneity For Binary Outcomes, Edward J. Mascha, Jeffrey M. Albert

Journal of Modern Applied Statistical Methods

Variability in individual causal effects, treatment effect heterogeneity (TEH), is important to the interpretation of clinical trial results, regardless of the marginal treatment effect. Unfortunately, it is usually ignored. In the setting of two-arm randomized studies with binary outcomes, there are estimators for bounds on the probability of control success and treatment failure for an individual, or the treatment risk. Here, those bounds were refined and the sampling properties were assessed using simulations of correlated multinomial data via the Dirichlet multinomial. Results indicated low bias and mean squared error. Moderate to high intraclass correlation (ICC) and large numbers of clusters …

Go to article

Two New Unbiased Point Estimates Of A Population Variance, Matthew E. Elam May 2006

Two New Unbiased Point Estimates Of A Population Variance, Matthew E. Elam

Journal of Modern Applied Statistical Methods

Two new unbiased point estimates of an unknown population variance are introduced. They are compared to three known estimates using the mean-square error (MSE). A computer program, which is available for download at http://program.20m.com, is developed for performing calculations for the estimates.

Go to article

Multiple Comparison Procedures, Trimmed Means And Transformed Statistics, Rhonda K. Kowalchuk, H. J. Keselman, Rand R. Wilcox, James Algina, James Algina, James Algina May 2006

Multiple Comparison Procedures, Trimmed Means And Transformed Statistics, Rhonda K. Kowalchuk, H. J. Keselman, Rand R. Wilcox, James Algina, James Algina, James Algina

Journal of Modern Applied Statistical Methods

A modification to testing pairwise comparisons that may provide better control of Type I errors in the presence of non-normality is to use a preliminary test for symmetry which determines whether data should be trimmed symmetrically or asymmetrically. Several pairwise MCPs were investigated, employing a test of symmetry with a number of heteroscedastic test statistics that used trimmed means and Winsorized variances. Results showed improved Type I error control than competing robust statistics.

Go to article

Confidence Intervals On Subsets May Be Misleading, Juliet Popper Shaffer May 2006

Confidence Intervals On Subsets May Be Misleading, Juliet Popper Shaffer

Journal of Modern Applied Statistical Methods

No abstract provided.

Go to article

The Effect On Type I Error And Power Of Various Methods Of Resolving Ties For Six Distribution-Free Tests Of Location, Bruce R. Fay May 2006

The Effect On Type I Error And Power Of Various Methods Of Resolving Ties For Six Distribution-Free Tests Of Location, Bruce R. Fay

Journal of Modern Applied Statistical Methods

The impact on Type I error robustness and power for nine different methods of resolving ties was assessed for six distribution-free statistics with four empirical data sets using Monte Carlo techniques. These statistics share an underlying assumption of population continuity such that samples are assumed to have no equal data values (no zero difference–scores, no tied ranks). The best results across all tests and combinations of simulation parameters were obtained by randomly resolving ties, although there were exceptions. The method of dropping ties and reducing the sample size performed poorly.

Go to article

Nonparametric Bayesian Multiple Comparisons For Dependence Parameter In Bivariate Exponential Populations, M. Masoom Ali, J. S. Cho, Munni Begum May 2006

Nonparametric Bayesian Multiple Comparisons For Dependence Parameter In Bivariate Exponential Populations, M. Masoom Ali, J. S. Cho, Munni Begum

Journal of Modern Applied Statistical Methods

A nonparametric Bayesian multiple comparisons problem (MCP) for dependence parameters in I bivariate exponential populations is studied. A simple method for pairwise comparisons of these parameters is also suggested. The methodology by Gopalan and Berry (1998) is extended using Dirichlet process priors, applied in the form of baseline prior and likelihood combination to provide the comparisons. Computation of the posterior probabilities of all possible hypotheses are carried out through a Markov Chain Monte Carlo, Gibbs sampling, due to the intractability of analytic evaluation. The process of MCP for the dependent parameters of bivariate exponential populations is illustrated with a numerical …

Go to article

Entropy Criterion In Logistic Regression And Shapley Value Of Predictors, Stan Lipovetsky May 2006

Entropy Criterion In Logistic Regression And Shapley Value Of Predictors, Stan Lipovetsky

Journal of Modern Applied Statistical Methods

Entropy criterion is used for constructing a binary response regression model with a logistic link. This approach yields a logistic model with coefficients proportional to the coefficients of linear regression. Based on this property, the Shapley value estimation of predictors’ contribution is applied for obtaining robust coefficients of the linear aggregate adjusted to the logistic model. This procedure produces a logistic regression with interpretable coefficients robust to multicollinearity. Numerical results demonstrate theoretical and practical advantages of the entropy-logistic regression.

Go to article

Comparison Of Some Simple Estimators Of The Lognormal Parameters Based On Censored Samples, Baklizi Ayman, Mohammed Al-Haj Ebrahem May 2006

Comparison Of Some Simple Estimators Of The Lognormal Parameters Based On Censored Samples, Baklizi Ayman, Mohammed Al-Haj Ebrahem

Journal of Modern Applied Statistical Methods

Point estimation of the parameters of the lognormal distribution with censored data is considered. The often employed maximum likelihood estimator does not exist in closed form and iterative methods that require very good starting points are needed. In this article, some techniques of finding closed form estimators to this situation are presented and extended. An extensive simulation study is carried out to investigate and compare the performance of these techniques. The results show that some of them are highly efficient as compared with the maximum likelihood estimator.

Go to article

Statistical Pronouncements V, Jmasm Editors May 2006

Statistical Pronouncements V, Jmasm Editors

Journal of Modern Applied Statistical Methods

No abstract provided.

Go to article

Properties Of The Gar(1) Model For Time Series Of Counts, Vasiliki Karioti, Chrys Caroni May 2006

Properties Of The Gar(1) Model For Time Series Of Counts, Vasiliki Karioti, Chrys Caroni

Journal of Modern Applied Statistical Methods

Models for time series count data include several proposed by Zeger and Qaqish (1988), subsequently generalized into the GARMA family. The GAR(1) model is examined in detail. The maximum likelihood estimation of the parameters will be discussed and the properties of Pearson and randomized residuals will be examined.

Go to article

Variance Estimation And Construction Of Confidence Intervals For Gee Estimator, Shenghai Zhang, Mary E. Thompson May 2006

Variance Estimation And Construction Of Confidence Intervals For Gee Estimator, Shenghai Zhang, Mary E. Thompson

Journal of Modern Applied Statistical Methods

The sandwich estimator, also known as the robust covariance matrix estimator, has achieved increasing use in the statistical literature as well as with the growing popularity of generalized estimating equations (GEE). A modified sandwich variance estimator is proposed, and its consistency and efficiency are studied. It is compared with other variance estimators, such as a model based estimator, the sandwich estimator and a corrected sandwich estimator. Confidence intervals for regression parameters based on these estimators are discussed. Simulation studies using clustered data to compare the performance of variance estimators are reported.

Go to article

Jmasm22: A Convenient Way Of Generating Normal Random Variables Using Generalized Exponential Distribution, Debasis Kundu, Anubhav Manglick May 2006

Jmasm22: A Convenient Way Of Generating Normal Random Variables Using Generalized Exponential Distribution, Debasis Kundu, Anubhav Manglick

Journal of Modern Applied Statistical Methods

A convenient method to generate normal random variable using a generalized exponential distribution is proposed. The new method is compared with the other existing methods and it is observed that the proposed method is quite competitive with most of the existing methods in terms of the K − S distances and the corresponding p-values.

Go to article

First Prev Next Last