This book was written by the architect of two MS Analytics programs and one undergraduate specialization in Business Analytics, with over a decade of experience teaching and practicing predictive analytics, and co-chairing premier academic conference mini-track in this field. The author's goal is to provide strong but understandable conceptual foundations and practical material for graduate students and managers, describing how to frame a business question, identify various model specification (i.e., feature engineering) and model methods (explainable and black box), select the optimal model based on the bias, variance, and cross-validation testing, and interpret results with meaningful storytelling for clients and managers. The book contains two components: (1) the main text with two sections—one with conceptual, mathematical, and managerial foundations, the other about advanced predictive modeling methods based on machine learning. The main text is further subdivided into two sections—Section 1 contains basic fundamentals of statistics and predictive modeling; Section 2 provides a deeper discussion of machine learning and advance predictive modeling approaches based on machine learning and cross-validation methods; and (2) a free appendix companion with annotated R Markdown code with hands-on applications, posted in GitHub.
