Statistics4MR-2022-23
Материал из Wiki - Факультет компьютерных наук
Версия от 18:04, 15 сентября 2022; Bdemeshev (обсуждение | вклад)
Содержание
Plan
Module 1
(1 lecture + 1 class per group X 7 weeks)
Bootstrap + Jacknife + CV Sampling Welch test Mann-Whitney test Sample size calculation Multiple comparison corrections
Module 2
(1 lecture + 1 class per group X 8 weeks)
CUPED Difference in Difference estimator Matching Contingency tables, Chi-squared tests ANOVA, ANCOVA Partial correlation + problem solving
Module 3
(1 lecture + 1 class per group X 11 weeks)
Discriminant analysis Logit PCA Factor analysis Cluster analysis, Dendrogramms Conjoint Analysis Multidimensional scaling
Grades
UoL grade = 0.7 * UoL exam + 0.3 * UoL coursework
Internal grades:
Fall grade = 0.4 * HA + 0.6 December Exam
Final grade = 0.5 * Fall grade + 0.2 HA + 0.3 * Final exam
Log-book
Week 1
Naive bootstrap, t-statistic bootstrap
Sources of wisdom:
Efron, Leisurely Look at bootstrap, jackknife and cv
Tim Hesterberg, What Teachers Should Know about the Bootstrap?
Week 2
Variance estimation: jackknife, bootstrap. Bootstrap in bootstrap.