Selected Reviews

“…this book delivered what the authors hoped to achieve—"Confidence Intervals for Discrete Data in Clinical Research." This book provides a comprehensive review of existing methods in constructing confidence intervals for binary and count data because these are the discrete data most frequently used in clinical research. Other types of discrete data, such as nominal data, multinominal data, or general categorical data, do not usually appear in clinical trials. In fact, the scope of application for this book can go beyond clinical trials. This book can also be a very useful reference tool for public health applications. Generally speaking, public health covers a much wider scope and includes many more practitioners. Most public health studies do not use randomization. Examples of these studies are epidemiology studies (including most observational studies), electronic medical records, registries, and medical data collected from insurance companies like claim databases, and so on. In other words, this book not only serves the readers it intended to serve but can also help potentially a much broader readership.” - N. Ting, Director of Biostatistics and Data Science at Boehringer Ingelheim Pharmaceuticals, Inc.   Biometrics, 79: 528-531. https://doi.org/10.1111/biom.13839

"For each method, they show plots of how actual coverage probabilities compare to a nominal level of 0.95 for 95% confidence intervals. Such plots are informative, showing that the Wald methods (i.e., maximum likelihood estimate plus and minus 1.96 estimated standard errors), which are the ones most commonly used in practice, usually perform poorly. .... All methods presented are easy enough to obtain these days with standard software. After introducing a method, the authors provide an example from a research study and discuss how to obtain the result using SAS or the excellent StatXact package for discrete data, with occasionalcitation of R functions." -Prof Alan Agresti, Department of Statistics, University of Florida, Gainesville, FL. Journal of American Statistical Asscociation, page 2945. https://doi.org/10.1080/01621459.2023.2262009

“In summary, the book by Pradhan and colleagues is a useful reference on the analysis of discrete data, mainly binomial and count data, for statisticians working in clinical research (and possibly beyond). They will appreciate the practical advice given and software implementations provided.” -Prof. T. Friede, Professor of Biostatistics, Department of Medical Statistics, University Medical Center Göttingen, Göttingen, Germany. Biometrical Journal, 65: 2300026. https://doi.org/10.1002/bimj.202300026