(Old Version)
STAT 4003
Statistical Inference (2019 Spring)
Class Information
- Class time: W 0830-1115
- Location: LSB LT6
- Outline
Instructor
- Name: Kin Wai Chan
- Email: kinwaichan@cuhk.edu.hk
- Office: LSB 115
- Tel: 3943 7923
- Office hour: I have an open door policy. Feel free to drop by anytime and ask me questions. (Of course, you may also make an appointment with me if you want a long meeting.)
Teaching Assistants
- Jiasheng Shi
- Email: 1155070160@link.cuhk.edu.hk
- Office: LSB G32
- Tel: 3943 8535
- Xiaoxiao Zhou
- Email: 1155113974@link.cuhk.edu.hk
- Office: LSB G22
- Tel: 3943 3048
Description
This course provides an introduction to statistical inference. Topics include statistical models, sampling distributions, asymptotic distributions, sufficiency, maximum likelihood estimation, Bayesian estimation, Rao-Blackwell theorem, Cramér-Rao theorem and the best unbiased estimator, Neyman-Pearson lemma, uniformly most powerful test and general likelihood ratio test.
Textbooks
A self-contained lecture note is the main source of reference. Complementary textbooks include
- (Major) Casella, G. and Berger, R. L. (2002). Statistical Inference. Duxbury Press.
- (Minor) Wasserman, L. (2013). All of statistics: a concise course in statistical inference. Springer.
Learning outcomes
Upon finishing the course, students are expected to
- apply basic statistics limit theorems and probability inequalities;
- perform data reduction through various kind of statistical principles;
- derive point estimation, interval estimation and testing procedures, and study their properties;
- analyze and evaluate statistical procedures by different criteria, and understand the interpretations of those criteria.
Assessment and Grading
The final score (out of 100) is given by {40a+20max(m,f)+40f}/100, where
- a (out of 100) is the average score of approximately eight assignments with the lowest two scores dropped;
- m (out of 100) is the score of mid-term exam; and
- f (out of 100) is the score of final exam.
Syllabus
- Probability theory
- (a) random variables, probability distribution, sampling distribution, statistics, exponential family
- (b) stochastic convergence, probability inequalities, statistics limit theorems
- (c) cumulant generating functions
- Data reduction
- (a) sufficiency principle
- (b) likelihood principle
- Point estimation
- (a) bias, variance, mean squared error, consistency
- (b) unbiased estimators, maximum likelihood estimators, minimax estimators, Bayes estimators, minimax estimators
- (c) (special topics) model misspecification, nuisance parameters, robustness, missing data, non-parametric estimation
- Hypothesis testing
- (a) power, size, types I/II errors, p-value
- (b) most powerful tests, Wald tests, Rao tests, likelihood ratio tests, permutation tests, goodness-of-fit tests
- (c) (special topics) false discovery rate, meta analysis,
- Interval estimation
- (a) coverage probability, expected width
- (b) duality, pivotal quantity, confidence intervals, credible intervals
- Discussion topics
- (a) comparison of Bayesian, frequentist, and fiducial (BFF) inference
- (b) causal inference
Lecture Notes (Draft)
Contents and Instructions
Part 1: Basic Probability and Statistics
Part II: Data Reduction
Part III: Point Estimation
Part IV: Hypothesis Testing
Part V: Interval Estimation
- Chapter 10: Quality of Interval Estimators
- Chapter 11: Methods of Interval Estimators
Appendix
- Appendix A: Basic Mathematics
Assignments
- Assignment 1: Stochastic convergence & statistical models (Due: 25 Jan (Fri) @ 6:00pm) Solution
- Assignment 2: Sufficiency principle (Due: 15 Feb (Fri) @ 6:00pm) Solution
- Assignment 3: Quality of point estimators & MoM Estimators (Due: 27 Feb (Wed) @ 6:00pm) Solution
- Assignment 4: UMVUE (Due: 20 Mar (Wed) @ 6:00pm) Solution
- Assignment 5: MLE (Due: 29 Mar (Fri) @ 6:00pm) Solution
- (Click here) Simulation demonstration and incomplete R-code for Exercise 5.2 (e)
- (Click here) Sample output of the simulation demonstration
- Assignment 6: UMPT (Due: 17 April (Wed) @ 6:00pm) Solution
- (Click here) Example of MPT and UMPT
- Assignment 7: LRT & asymptotic tests (Due: 26 April (Fri) @ 6:00pm) Solution
- Assignment 8: Confidence intervals (** Optional ** No need to hand in) Solution
Mid-term Exam
- Date: 6 March (Wednesday)
- Time: ~ 9:00 am (A short review will be given before the mid-term exam).
- Duration: 80 minutes
- Scope: Chapter 1 - Chapter 7 page 15 (i.e., everything until UMVUE)
- Remarks:
- One piece of double-sided A4-size hand-written cheat sheet is allowed.
- No calculator is allowed.
- No make-up mid-term exam (except for those earlier approved cases) will be given.
- Mock mid-term exam and Solution
- Mid-term exam and Solution
Final Exam
- Date: 2 May (Thursday)
- Time: 1530 - 1730
- Duration: 120 minutes
- Location: University Gymnasium
- Scope: Chapter 1 - Chapter 11 (i.e., everything taught in this course)
- Remarks:
- One piece of double-sided A4-size hand-written cheat sheet is allowed.
- No calculator is allowed.
- Mock final exam and Solution (Typos: Question 1.3-1.4, Question 3 (5))
- Final exam and Solution
Review Lecture
- Date: 24 April (Wednesday)
- Time: 1030 - 1230
- Location: LSB LT6
- Remarks:
- I will go through some examples in this lecture.
- If time allows, some extra materials will also be taught.