Computationally Intensive Statistics
The term "computationally intensive statistics" covers a very broad range of topics. I will attempt to give a brief overview of the field, touching on topics such as modern Bayesian statistics, multiple imputation, and neural network modeling. However, we will focus on what has arguably been the most important application of brute-force computing power to statistical analysis, namely resampling methods, especially the use of the bootstrap for error estimation. In his original book on the subject, Brad Efron said that the mathematical level was easy, but "The statistical ideas run deep, sometimes over our head at our current level of understanding." I will try to explain the methods in simple terms, and to demonstrate why Efron's statement is (still) true.
Fee: None to members of the Cornell community, but registration is required. Since space is limited, early registration is encouraged.
Instructor: Professor Bunge, ILR Social Statistics
For times and locations of upcoming workshops, please see the Workshop Schedule.