Bayesian Paleoethnobotany Styles of Data Analysis The Data Research Plan Prior Probabilities of Activities Botanical Remains: The Likelihood Function Computation of Posterior Distributions Conclusion References Discussion Reply to the Discussion Epilogue Teaching Suggestions Statistical Sampling in Tax Audits Foreword Published Paper Introduction The Theory of Statistical Sampling: An Overview Judicial Decisions The Case for Statistical Samples Statutory Standards for the Use of Statistical Sampling Conclusions Appendix A Epilogue Teaching Suggestions A Statistical Analysis of Adverse Impact of Employer Decisions (1990) 53 Foreword Published Paper Introduction Database Issues Analyses Fisher's Exact Tests A Bayesian Analysis Paired Observations Conclusions References Epilogue Teaching Suggestions Subjective Bayesian Analysis for Surveys with Missing Data Foreword Published Paper Introduction Background and Analysis Conclusions References Epilogue Teaching Suggestions Missing Data in the Forensic Context Foreword Published Paper Introduction The New Jersey Turnpike Case Presentation to Court of Surveys with Missing Data References Epilogue Teaching Suggestions Bayesian Demography: Projecting the Iraqi Kurdish Population Foreword Published Paper Introduction Current Practices The Iraqi Kurdish Population Overview of Projection Procedure Projecting the Iraqi Kurdish Population Fertility Mortality Migration Obtaining a Base Population Results Discussion References Epilogue Teaching Suggestions An Allegation of Examination Copying Foreword Pragmatics of Uncertainty Published Paper Data and Formal Analysis Priors: What to Look For An Index Results Conclusion Other Purposes, Other Indices References Epilogue Teaching Suggestions Vote Tampering in a District Justice Election in Beaver County, PA Foreword Published Paper Introduction Legal History Model Data Structuring Model Speci cation Estimation with Markov Chain Monte Carlo Results Model Validation Discussion Appendix References Epilogue Teaching Suggestions The Effect of Intensity of Effort to Reach Survey Respondents: A Toronto Smoking Survey Foreword Published Paper Introduction The Survey Overview of Methodology MAR Models Logistic Regression Bayesian Logistic Regression NI Model Modeling the Non-Response Mechanism Relating Non-Response to the Dependent Variable - The NI Model 166 Data Augmentation Sampling from the Posterior Distribution Choice of Prior Distributions Results Coe_cients for the Number of Calls Sensitivity to Priors Effect on Odds of Response Effect on Probability of Response Conclusion Appendices References Epilogue Teaching Suggestions Comparing Harm Done by Mobility and Class Absence: Missing Students and Missing Data Foreword Published Paper Introduction Data Data Description Missing Data Model Specification Notation Likelihood without Choice Indicators Likelihood with Choice Indicators Prior Distributions Methodology for Sampling from the Posterior Distribution Results Conclusions Discussion Appendix References Editor's Foreword Comment Epilogue Teaching Suggestions Hierarchical Models for Employment Decisions Foreword Published Paper Introduction Proportional Hazards Models Hierarchical Priors for Time Varying Coe_cients Smoothness Priors Forming an Opinion about Smoothness Posterior Distribution Reexpressing Examples Case K Case W Valentino v. United States Postal Service Discussion References Epilogue Teaching Suggestions Pragmatics of Uncertainty Age- and Time-Varying Proportional Hazards Models for Employment Discrimination Foreword Published Paper Introduction Proportional Hazards Regression Overview Proportional Hazards Models for Time to Event Data Thin-Plate Spline Smoothness Priors Case W Revisited Forming an Opinion about Smoothness and Anisotropy Computing the Posterior Distribution Sensitivity Analysis Identi cation of the Anisotropy Parameter Previous Analyses of Case W Summary Discussion References Epilogue Teaching Suggestions Error Analysis for Small Angle Neutron Scattering Datasets Using Bayesian Inference Foreword Published Paper Introduction Correlation Lengths Small-angle Neutron Scattering The SANS Instrument Instrument Components Instrument Con gurations SANS Error Analysis Traditional Error Analysis The Bayesian Alternative Results and Discussion Separate Marginal Distributions on _L and _S Joint Posterior Distribution on (_L; _S) Representing Fits Graphically: family of curves" and Standardized Residual Distributions Conclusions Appendix A Appendix B References Comment Comment Rejoinder Epilogue Teaching Suggestions Impacts of Beliefs about Tropical Cyclone Detection on Conclusions about Trends in Tropical Cyclone Number Foreword Published Paper Introduction TC Detection and Recording over the Years Combining Detection Probability with Raw Counts Illustration Sensitivity to Prior Quanti cation Concluding Remarks Resolving Ambiguity with Further Studies Sensitivity to Formulation Toward a Flexible Formulation Incorporating Climate Models References Epilogue Teaching Suggestions The Number of Killings in Southern Rural Norway Foreword Published Paper Norwegian Homicide Law and the Documentary Evidence Demographic Evidence about the Number of Killings Models of the Data Conclusion References Epilogue Teaching Suggestions Epilogue Index
Dr. Polhemus is Chief Technology Officer for Statpoint Technologies, Inc. and directs the development of STATGRAPHICS. He received his B.S.E. and Ph.D. degrees from the School of Engineering and Applied Science at Princeton University, under the tutelage of Dr. J. Stuart Hunter. Dr. Polhemus spent two years as an assistant professor in the Graduate School of Business Administration at the University of North Carolina at Chapel Hill and six years as an assistant professor in the Engineering School at Princeton University.Dr. Polhemus founded Statistical Graphics Corporation in 1980 to develop and promote the STATGRAPHICS software program. In 1983, he founded Strategy Plus, Inc., which developed EXECUSTAT for DOS. Dr. Polhemus founded NWP Associates, Inc., in 1993 to develop STATLETS, a set of Java applets which permit statistical data analysis over the Internet. In 1999 development of STATGRAPHICS was assumed by Statpoint Technologies, Inc.Dr. Polhemus lives in northern Virginia with his wife Caroline and is the proud father of four sons: Christopher, Gregory, Leland, and Michael.