Course that focuses on topical subjects in specific fields of neuroscience.

This is an intermediate-level statistics class geared toward PhD students in neuroscience and related fields. This is a cooperatively run course in which students take the lead in collaboratively learning, presenting material, and doing hands-on exercises, under the guidance of a GSI with assistance from a faculty member. Students will come away from this class understanding the concepts and practice of commonly-used statistical and data modeling methods in neuroscience, when to use different methods, what assumptions they make, and how they relate to one another. Students learn the caveats and limitations of various methods, and the role of statistical analysis in the larger process of good experimental design. The first half of the course covers traditional statistics in a moderate level of detail (descriptive statistics, data visualization, t-tests, ANOVA, correlation, non-parametric tests, and bootstrapping). The second half of the course covers concepts in modeling of data (specifying, fitting, and evaluating models, with a focus on the general linear model), as well as resampling methods, classification, and clustering. Weekly exercises teach students how to implement statistical methods in Python, using real data.

Prerequisites: Students are expected to have an undergraduate introductory statistics course (psych stats, bio stats) or equivalent. If you don’t have this background, you should be able to catch up with extra readings each week. No programming experience is required.

Repeat Rules