Papers by Anthony Rossini
A Semiparametric Proportional Odds Regression Model for the Analysis of Current Status Data
... AA Tsiatis is Professor, Department of Biostatistics, Harvard School of Public Health, Boston... more ... AA Tsiatis is Professor, Department of Biostatistics, Harvard School of Public Health, Boston, MA 02115. ... Quantiles of Standard Normal Figure 1. QQ Plots for the Distribution of 400 Score Statistics From the Exponential Model, 75% Left Censored, With 50 Observations per Data ...
Jaids Journal of Acquired Immune Deficiency Syndromes, Mar 1, 2005
Background: In surveys, clients have expressed preferences for alternatives to traditional HIV co... more Background: In surveys, clients have expressed preferences for alternatives to traditional HIV counseling and testing. Few data exist to document how offering such alternatives affects acceptance of HIV testing and receipt of test results. Objectives: This randomized controlled trial compared types of HIV tests and counseling at a needle exchange and 2 bathhouses to determine which types most effectively ensured that clients received test results. Methods: Four alternatives were offered on randomly determined days: (1) traditional test with standard counseling, (2) rapid test with standard counseling, (3) oral fluid test with standard counseling, and (4) traditional test with choice of written pretest materials or standard counseling. Results: Of 17,010 clients offered testing, 7014 (41%) were eligible; of those eligible, 761 (11%) were tested: 324 at the needle exchange and 437 at the bathhouses. At the needle exchange, more clients accepted testing (odds ratio [OR] = 2.3; P < 0.001) and received results (OR = 2.6; P < 0.001) on days when the oral fluid test was offered compared with the traditional test. At the bathhouses, more clients accepted oral fluid testing (OR = 1.6; P < 0.001), but more clients overall received results on days when the rapid test was offered (OR = 1.9; P = 0.01). Conclusions: Oral fluid testing and rapid blood testing at both outreach venues resulted in significantly more people receiving test results compared with traditional HIV testing. Making counseling optional increased testing at the needle exchange but not at the bathhouses.
International Journal of Parallel Programming, 2009
This paper presents a simple parallel computing framework for the statistical programming languag... more This paper presents a simple parallel computing framework for the statistical programming language R. The system focuses on parallelization of familiar higher level mapping functions and emphasizes simplicity of use in order to encourage adoption by a wide range of R users. The paper describes the design and implementation of the system, outlines examples of its use, and presents some possible directions for future developments.
Workstations and software for statistical analysis: Some thoughts on environmental concerns
Literate Statistical Practice (LSP) is an method for statistical practice which suggests that doc... more Literate Statistical Practice (LSP) is an method for statistical practice which suggests that documentation and specification occur at the same time as statistical coding. It applies literate programming Knuth (1992) to the practice of statistics. We discuss 2 different approaches for LSP, one currently implemented using Emacs with Noweb and Emacs Speaks Statistics (ESS), and the other developed based on eXtensible Markup Language (XML) tools. The interference needed to change an individual's habits comes at a high cost, and good tools are critical for encouraging people to switch to a more literate style. We discuss why LSP can help, and suggest how ESS can ease the burden placed on the analyst.
Statistical Tools on the World Wide Web
Amer Statist, 1998
The World Wide Web (WWW) represents a powerful tool for furthering the development and practice o... more The World Wide Web (WWW) represents a powerful tool for furthering the development and practice of statistics. The GASP (Globally Accessible Statistical Procedures) WWW site has been set up as a primary listing of statistical procedures which can be used over the WWW. This article highlights several possible approaches for making a procedure WWW accessible. These approaches effectively solve many
Sampling Designs for HIV Molecular Epidemiology with Application to Honduras
Http Dx Doi Org 10 1089 Aid 2005 21 907, Dec 1, 2005
Proper sampling is essential to characterize the molecular epidemiology of human immunodeficiency... more Proper sampling is essential to characterize the molecular epidemiology of human immunodeficiency virus (HIV). HIV sampling frames are difficult to identify, so most studies use convenience samples. We discuss statistically valid and feasible sampling techniques that overcome some of the potential for bias due to convenience sampling and ensure better representation of the study population. We employ a sampling design called stratified cluster sampling. This first divides the population into geographical and/or social strata. Within each stratum, a population of clusters is chosen from groups, locations, or facilities where HIV-positive individuals might be found. Some clusters are randomly selected within strata and individuals are randomly selected within clusters. Variation and cost help determine the number of clusters and the number of individuals within clusters that are to be sampled. We illustrate the approach through a study designed to survey the heterogeneity of subtype B strains in Honduras.
Literate Statistical Practice (LSP) is an method for statistical practice which suggests that doc... more Literate Statistical Practice (LSP) is an method for statistical practice which suggests that documentation and specification occur at the same time as statistical coding. It applies literate programming Knuth (1992) to the practice of statistics. We discuss 2 different approaches for LSP, one currently implemented using Emacs with Noweb and Emacs Speaks Statistics (ESS), and the other developed based on eXtensible Markup Language (XML) tools. The interference needed to change an individual's habits comes at a high cost, and good tools are critical for encouraging people to switch to a more literate style. We discuss why LSP can help, and suggest how ESS can ease the burden placed on the analyst.
Ozone concentrations exhibit spatial variability within metropolitan areas, resulting in signific... more Ozone concentrations exhibit spatial variability within metropolitan areas, resulting in significantly different personal exposures among individuals. This paper uses the statistical technique, kriging, to explore the 12-h daytime (Sam-8pm) ozone spatial variation and to predict mean outdoor ozone levels at home sites within the Toronto metropolitan area. Outdoor ozone measurements taken in the Toronto metropolitan area between June and August 1992 are used in kriging models to predict outdoor ozone concentrations. The performance of the model is evaluated by comparing actual home outdoor measurements with predicted values. Results indicate that kriging predictions are more accurate than using only the closest stationary ambient site measurements for determining home outdoor ozone concentrations within the metropolitan area. The average variogram obtained from pooling data throughout the entire sampling period shows a clear spatial trend in the outdoor ozone variation. Kriging predictions using the parameters from the average variogram perform as well as those using variograms from individual days. An approach for minimizing sample bias can be used to increase the accuracy of the predictions; cross-validation suggests that it is a reasonable procedure. C~>p.vr;ghr aIYY6 El.~ierSciem Lid
Overcoming barriers to HIV testing: preferences for new strategies among clients of …
JAIDS Journal of …
This study was supported by a grant from the University of Washington Center for AIDS Research Ne... more This study was supported by a grant from the University of Washington Center for AIDS Research New Investigator Award No. AI27757 and from cooperative agreement No. R18/CCR015258-01 with the Centers for Disease Control and Prevention. Time to ...
Phase II Oncology Trials: Perspective from Industry
Brown/A Practical, 2014
Modeling and Simulation in Clinical Drug Development
Drug Discovery, Development, and Manufacturing, 2010
4Modeling and Simulation, Novartis Institutes for BioMedical Research, Inc., Cambridge, Massachus... more 4Modeling and Simulation, Novartis Institutes for BioMedical Research, Inc., Cambridge, Massachusetts ... 1 Introduction 1.1 What Is a Model? 1.2 What Is Simulation? 1.3 Model Building 1.4 Bayesian Methods 1.5 What Lies Ahead 2 Types of Modeling and Simulation 2.1 Regression ...
Manni, A. et al. Induction of a less aggressive breast cancer phenotype by protein kinase C- and -overexpression. Cell Growth Differ. 7, 1187-1198
Cell growth & differentiation: the molecular biology journal of the American Association for Cancer Research
ABSTRACT
AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium
This paper presents the implementation of a model for expression array annotation (EAA) using the... more This paper presents the implementation of a model for expression array annotation (EAA) using the BioMediator biological data integration system along with BioConductor, an analytic tools platform. The model presented addresses the need for annotation sources identified during BioConductor's development. Annotation provides us with well-curated genomic background knowledge for expression array analysis and interpretation. Annotation requests are constructed and posted to the query interface of the EAA package (the EAA model implemented as a component of BioConductor). The software enumerates all possible annotation paths for queries. These are then transformed to PQL queries and processed by BioMediator. Annotation entities returned from the EAA package answer the annotation request.
Computable Statistical Research and Practice
COMPSTAT 2008, 2008
ABSTRACT The work of research and applied statisticians is driven by both manual and electronic c... more ABSTRACT The work of research and applied statisticians is driven by both manual and electronic computation. This computation, which supports our work, can be crudely described by 3 stages: scoping, to assess what can and needs to be done; analysis, where this is accomplished; and reporting, which communicates the results to others. Barriers to the reuse of computations can be found in the translational needs driving the transition between sub-activities; for example, scoping activities are seldom directly reusable during analysis, and there is a limited amount of direct reuse going from analysis to reporting. There is an additional high barrier for translating statistical theory and methodology to practical activities, with both sides (applied and theoretical statisticians) pointing the blame at the other for not using appropriate tools or addressing appropriate needs. In this sense, much statistical research is not really computable, but rather, translatable. This work describes some proposals for exploring novel information technology support to address the translational challenges during transition between stages of statistical practice. These are illustrated in the on-going design of CommonLisp Stat, a platform for investigating the interaction between statistical computing environments supporting research and/or practice.
A Semiparametric Proportional Odds Regression Model for the Analysis of Current Status Data
Journal of The American Statistical Association, 1996
An approach to covariate analysis of current status data is described. The method is applicable w... more An approach to covariate analysis of current status data is described. The method is applicable when the logit of the conditional probability of survival given the covariates is some increasing function of time plus a linear combination of the covariates. The approach is based on approximating the infinite-dimensional nuisance parameter, the baseline log-odds of failure, with a step function, and
Statistical Tools on the World Wide Web
American Statistician, 1998
The World Wide Web (WWW) represents a powerful tool for furthering the development and practice o... more The World Wide Web (WWW) represents a powerful tool for furthering the development and practice of statistics. The GASP (Globally Accessible Statistical Procedures) WWW site has been set up as a primary listing of statistical procedures which can be used over the WWW. This article highlights several possible approaches for making a procedure WWW accessible. These approaches effectively solve many
Book Review¶“Applied Smoothing Techniques for Data Analysis: The Kernel Approach with S-Plus Illustrations” by Adrian W. Bowman and Adelchi Azzalini
Computational Statistics, 2000
Workstations and software for statistical analysis: Some thoughts on environmental concerns
Uploads
Papers by Anthony Rossini