Search

A special-topics undergraduate course about the history of statistics which was taught in Spring 2023 at the University of South Carolina is described. We review other similar courses (past and current) and explain the discussion-based nature of this course. The conception and planning of the course are detailed, and the unique experiences (activities, guest speakers, presentations, etc.) are described. The course emphasized substantial amounts of independent reading outside of class and lively discussions during class. Topics covered in the class include the early development of probability, the normal distribution, and the central limit theorem; the development of modern statistical science by British statisticians; the rise of formal mathematical statistics; and increasing specialization and modern computational and data-analytic advances. An assessment of the course's effectiveness based on qualitative student survey data is given. Students were highly complimentary of the course, praising the in-class discussion format, the benefits of doing the outside readings, the invited guest speakers, and the in-class activities. There were occasional comments that the amount of required reading was excessive. Based on this, suggestions for future offerings of the course are presented, including developing a more carefully curated set of readings. [ABSTRACT FROM AUTHOR]

AbstractDistance covariance (Székely et al. 2007) is a fascinating recent notion, which is popular as a test for dependence of any type between random variables X and Y. This approach deserves to be touched upon in modern courses on mathematical statistics. It makes use of distances of the type |X−X′| and |Y−Y′| , where (X′,Y′) is an independent copy of (X,Y) . This raises natural questions about independence of variables like X−X′ and Y−Y′ , about the connection between Cov(|X−X′|,|Y−Y′|) and the covariance between doubly centered distances, and about necessary and sufficient conditions for independence. We show some basic results and present a new and nontechnical counterexample to a common fallacy, which provides more insight. We also show some motivating examples involving bivariate distributions and contingency tables, which can be used as didactic material for introducing distance correlation. [ABSTRACT FROM AUTHOR]