Statistical Challenges with Big Data in Management Science

In the past few years, there has been an increasing awareness that the enormous amount of data being captured by both public and private organisations can be profitably used for decision making. Aided by low-cost computer hardware, fast processing speeds and advancements in data storage technologies, Big Data Analytics has emerged as a fast growing field. However, the statistical challenges that are faced by statisticians and data scientists, while doing analytics with Big Data has not been adequately discussed. In this paper, we discuss the several statistical challenges that are encountered while analyzing Big data for management decision making. These challenges give statisticians significant opportunities for developing new statistical methods. Two methods—Symbolic Data Analysis and Approximate Stream Regression—which holds promise in addressing some of the challenges with Big Data are discussed briefly with real life examples. Two case studies of applications of analytics in management—one in marketing management and the other in human resource management—are discussed.