Current Search: Social sciences  Statistical methods (x)
View All Items
 Title
 The Use Of Overlapping Vs. NonOverlapping Moving Block Bootstrapping To Estimate The Variance Of A Statistic Of Dependent Data.
 Creator
 Davis, Benjamin F., Radulovic, Dragan, Florida Atlantic University
 Abstract/Description

Determining the variance of a statistic (such as the sample median) can be difficult. Various methods of Bootstrapping (resampling with replacement) were used to estimate variance of one or more statistics based on a single sample. This estimator was compared to the empirical estimators based on repeated simulations of various sample sizes from a given distribution. Of particular interest was which of the methods of Bootstrapping were most effective with a dependent data set. Different...
Show moreDetermining the variance of a statistic (such as the sample median) can be difficult. Various methods of Bootstrapping (resampling with replacement) were used to estimate variance of one or more statistics based on a single sample. This estimator was compared to the empirical estimators based on repeated simulations of various sample sizes from a given distribution. Of particular interest was which of the methods of Bootstrapping were most effective with a dependent data set. Different degrees of dependency were used for the simulations with dependent data.
Show less  Date Issued
 2008
 PURL
 http://purl.flvc.org/fau/fd/FA00000738
 Subject Headings
 Bootstrap (Statistics), Social sciencesStatistical methods, Mathematical statistics, Sampling (Statistics), Estimation theory, Nonparametric statistics
 Format
 Document (PDF)
 Title
 Variances in metaanalysis.
 Creator
 Humphreys, Katherine L. B., Florida Atlantic University, Qian, Lianfen
 Abstract/Description

Metaanalysis is a statistical method of combining many individual analyses. This thesis reviews the need for metaanalysis; the many statistical consideration facing the metaanalyst; and some of Hedges' results concerning the combined estimate of effect size with unequal weights from his 1981 and 1982 papers. Unequal weights used to combine estimates of effect size in metaanalysis are derived using the variances given by the large sample, normal approximation of the distribution of Hedges'...
Show moreMetaanalysis is a statistical method of combining many individual analyses. This thesis reviews the need for metaanalysis; the many statistical consideration facing the metaanalyst; and some of Hedges' results concerning the combined estimate of effect size with unequal weights from his 1981 and 1982 papers. Unequal weights used to combine estimates of effect size in metaanalysis are derived using the variances given by the large sample, normal approximation of the distribution of Hedges' unbiased estimates of effect sizes. These variances depend on the effect size and the sample sizes of both experimental and control groups. This creates circular definitions and calls for further estimates. This thesis analyzes the limiting normal approximation to derive a variance which is not dependent on effect size, and it provides guidelines for its use.
Show less  Date Issued
 1998
 PURL
 http://purl.flvc.org/fcla/dt/15544
 Subject Headings
 Metaanalysis, Social sciencesStatistical methods
 Format
 Document (PDF)
 Title
 Discounting the role of causal attributions in the ANOVA model of attribution.
 Creator
 Hakala, Kori A., Charles E. Schmidt College of Science, Department of Psychology
 Abstract/Description

For years attribution research has been dominated by the ANOVA model of behavior which proposes that people construct their dispositional attributions of others by carefully comparing and weighing all situational information using mental computations similar to the processes used by researchers to analyze data. A preliminary experiment successfully determined that participants were able to distinguish differences in variability assessed across persons (high vs. low consensus) and across...
Show moreFor years attribution research has been dominated by the ANOVA model of behavior which proposes that people construct their dispositional attributions of others by carefully comparing and weighing all situational information using mental computations similar to the processes used by researchers to analyze data. A preliminary experiment successfully determined that participants were able to distinguish differences in variability assessed across persons (high vs. low consensus) and across situations (high vs. low distinctiveness). Also, it was clear that the subjects could evaluate varying levels of situational constraint. A primary experiment administered to participants immediately following the preliminary study determined that participants grossly underutilized those same variables when making dispositional attributions. Results gave evidence against the use of traditional ANOVA models and support for the use of the Behavior Averaging Principle of Attribution.
Show less  Date Issued
 2008
 PURL
 http://purl.flvc.org/FAU/166450
 Subject Headings
 Social sciences, Statistical methods, Analysis of variance, Data processing, Mathematical statistics, Attribution (Social psychology)
 Format
 Document (PDF)
 Title
 Sensitivity analysis of predictive data analytic models to attributes.
 Creator
 Chiou, James, Zhu, Xingquan, Florida Atlantic University, College of Engineering and Computer Science, Department of Computer and Electrical Engineering and Computer Science
 Abstract/Description

Classification algorithms represent a rich set of tools, which train a classification model from a given training and test set, to classify previously unseen test instances. Although existing methods have studied classification algorithm performance with respect to feature selection, noise condition, and sample distributions, our existing studies have not addressed an important issue on the classification algorithm performance relating to feature deletion and addition. In this thesis, we...
Show moreClassification algorithms represent a rich set of tools, which train a classification model from a given training and test set, to classify previously unseen test instances. Although existing methods have studied classification algorithm performance with respect to feature selection, noise condition, and sample distributions, our existing studies have not addressed an important issue on the classification algorithm performance relating to feature deletion and addition. In this thesis, we carry out sensitive study of classification algorithms by using feature deletion and addition. Three types of classifiers: (1) weak classifiers; (2) generic and strong classifiers; and (3) ensemble classifiers are validated on three types of data (1) feature dimension data, (2) gene expression data and (3) biomedical document data. In the experiments, we continuously add redundant features to the training and test set in order to observe the classification algorithm performance, and also continuously remove features to find the performance of the underlying classifiers. Our studies draw a number of important findings, which will help data mining and machine learning community under the genuine performance of common classification algorithms on realworld data.
Show less  Date Issued
 2014
 PURL
 http://purl.flvc.org/fau/fd/FA00004274, http://purl.flvc.org/fau/fd/FA00004274
 Subject Headings
 Data mining, Forecasting  Mathematical models, Social sciences  Statistical methods, Ubiquitous computing
 Format
 Document (PDF)