Based on a starter course for beginning graduate students, Core Statistics provides concise coverage of the fundamentals of inference for parametric statistical models, including both theory and practical numerical computation. The book considers both frequentist maximum likelihood and Bayesian stochastic simulation while focusing on general methods applicable to a wide range of models and emphasizing the common questions addressed by the two approaches. This compact package serves as a lively introduction to the theory and tools that a beginning graduate student needs in order to make the transition to serious statistical analysis: inference; modeling; computation, including some numerics; and the R language. Aimed also at any quantitative scientist who uses statistical methods, this book will deepen readers' understanding of why and when methods work and explain how to develop suitable methods for non-standard situations, such as in ecology, big data and genomics.
In The Press
'The author keeps this book concise by focusing entirely on topics that are most relevant for scientific modeling via maximum likelihood and Bayesian inference. This makes it an ideal text and handy reference for any math-literate scientist who wants to learn how to build sophisticated parametric models and fit them to data using modern computational approaches. I will be recommending this well-written book to my collaborators.' Murali Haran, Pennsylvania State University