Statistical Methods for Neuroscience and Psychology36-746 Spring, 2014Schedule: TTh 10:30–11:50 Instructor: Rob KassCNBC Classroom MI 130 Baker Hall 229HFirst meeting: Jan. 14 [email protected], 268-8723This course provides a brief survey of statistical methods that are of use inneuroscience and psychology. It combines the material from a first coursein statistics with some important advanced topics. No specific backgroundin statistics is assumed other than basic statistical ideas as taught in highschool, and perhaps a bit of name-recognition of a small number of widely-used methods. The mathematical prerequisites are familiarity with vectorsand matrices; the exponential and log functions; and elementary calculus. Iwill go over all statistical methods from first principles, discussing why theyare used and how their results should be interpreted. The pace will be abouttwice as fast as a standard basic statistics course. A key goal of the course isto have students begin to understand what it is like to “think statistically.”The text for the course will be my forthcoming book Analysis of Neural Data,written with co-authors Emery Brown (MIT/MGH) and Uri Eden (BostonU.). A pre-copy-editing draft of individual chapters will be posted on thecourse website. We will go through Chapters 1–13 and selected portions ofChapters 14-19, depending in part on the interests of students.Auditors are welcome. Registered students will be required to read the as-signed sections of the text and post a comment or question on the course blogprior to 9:30 am preceding the 10:30 class, so that I have time to look overall the comments immediately before class. The comments will be privateuntil I read them, at which time I will make them public and I will attemptto address all the points raised during the lecture that day. There will alsobe many short-answer assignments on the material, and approximately 4 re-quired data analysis assignments. Some instruction in using the softwareMatlab and R will be provided outside of class, by the Course AssistantPengcheng Zhou ([email protected]). There will be no exams.1Please note that I will march through these chapters roughly in the orderthey are written, but will deviate by jumping back and forth a bit in orderto bring some of the more applicable material in earlier, as follows:Part I (Elementary Statistics): Chs. 1–7, 10, 12.1–12.4, 13.1.Part II (Basic Statistical Theory): Chs. 8, 9, 11, 12.5, 13.2–13.4.Part III (Advanced Topics): Selections from Chs. 14–19.Parts I and II will take approximately 12 classes and 7 classes, respectively.Analysis of Neural Data: Contents1. Introduction2. Exploring Data3. Probability and Random Variables4. Random Vectors5. Important Probability Distributions6. Sequences of Random Variables7. Estimation and Uncertainty8. Estimation in Theory and Practice9. Propagation of Uncertainty and the Bootstrap10. Models, Hypotheses, and Statistical Significance11. General Methods for Testing Hypotheses12. Linear Regression13. Analysis of Variance14. Generalized Linear and Nonlinear Regression15. Bayesian Methods16. Nonparametric Regression17. Multivariate Analysis18. Time Series19. Point ProcessesAppendix: Mathematical Background2Grading PolicyGrades will be allocated, approximately, as follows:• Participation in commentaries on reading, 25%.• Short-answer homework, 25%.• Data analysis homework, 50%.We reserve the right to grade tardy assignments as
or
We will never post anything without your permission.
Don't have an account? Sign up