Stat 153. Main problem of this course is that I do not really know the material. The advanced material is okay because it has a solid theoretical basis. But the early stuff is ad hoc and case dependent – the worst type of material to learn. Also the students will need experimental data to explore the models – ideally data that corresponds to realistic situations, so where do I get this from? Also there is the infamous R – though probably it is okay if Jose gives them a basic introduction and they use the manual for the rest. Material to be covered. Lectures 1 & 2: Chapter 1 of Little & Rubin. This seems fairly straightforward. Though it contains a lot of case by case study. Main points – there is missing data. Data a_{ij}. Index m_{ij} Different ways that data can be missing. Mechanisms that lead to missing data. Missing Completely at Random(MCAR). Missing at Random (MAR). Taxonomy of Missing Data Methods. Probably enough for two lectures. Lecture 3 : Chapter 2 of Little & Rudin. Discussion of Least Squares and Yates Method. Estimate quantities and the variance of the estimate. Probably one lecture. Lecture 4: Chapter 3 of Little & Rudin Complete-Case and Available-Case analysis. I do not really understand the weighting methods. They seem ad hoc. Probably one lecture. Lecture 5: Chapter 4 of Little & Rudin. Single Imputation Methods – fill in the missing data. Mean/Regression/Stochastic regression imputation. Lecture 6: Chapter 5 of Little & Rudin. Estimates of Imputation Uncertainty. Bootstrap & Jacknife. Lecture 7 & 8: Chapter 6 of Little & Rudin. Likelihood Functions Methods. Review of standard ML methods (one lecture). Extension to cases with missing data. (MAR) versus non-MAR. Lecture 9 & 10 Chapter 8 of Little & Rudin. Expectation-Maximization Algorithm + Examples. Lecture 11. Chapter 9 of Little & Rudin. Standard errors based on the information matrix. Lecture 12 & 13. Chapter 10 of Little & Rudin. Data Augmentation and Mulitple Imputation. Lecture 14, 15, 16. Examples. Chapters 11 –12. Lecture 17, 18. Pearl’s work. Lecture 19. Review.