Download as pdf or txt
Download as pdf or txt
You are on page 1of 2

Recommended revision questions

• Week 1: Questions: 5, 7
• Week 3: Questions: 4, 6, 14, 18
• Week 4: Questions: 6, 7, 12, 18
• Week 5: Questions: 1, 8, 11, 13
• Week 7: Questions: 4, 6, 7, 14
• Week 8: Questions: 4, 8, 9
• Week 9: Questions: 1, 2, 13
• Week 10: Questions: 4, 7, 9
• Week 10 - Extra: Questions: 3, 5, 8

Probability
Basic Probability

• Basic discrete probability (Do not worry about Hyper-geometric distribution)


• Bayes formula
• Conditional probability and independence

Common Distributions

• Computing the moments and moment generating function of random variables.


• Computing the distribution of simple transformed random variables. Essentially Linear,
Parabolic and Square root.
• Bivariate density transform and doing double integrals.
• Computing sum of random variables using convolution formula, bivariate transform method
and moment generating functions, especially Gamma and Normal random variables.
• There could be a question using the Beta distribution?!.

Inequalities

• Jensen’s inequality
• Markov and Chebyshev’s inequality.

Limit of random variables

• Central Limit Theorem


• Convergence in Distribution
• Convergence in Probability
• Law of Large numbers
• Statement and application of Delta Method.

1
Statistical Inference
Estimators:

• Computing Method of Moment Estimators, MLE estimators.


• Method of evaluating estimators. i.e. Bias, Variance, Standard Error and MSE of an estimator.
• Compute the Fisher Information for the MLE. (No question on bivariate MLEs)
• Know the properties of the MLE.

– Asymptotically normal.
– The variance of the MLE estimator is approximately the inverse of the Fisher information.
– Invariance property.

Confidence intervals and Hypothesis testing:

• Given an estimator θb of a parameter θ, understand the procedure of constructing a (1 − α)%


confidence interval and study the examples.
• Construction of approximate confidence intervals using MLEs (Similar to the normal case).

• Understand the procedure in doing a hypothesis test. Be able to do the Asymptotic/Approximate


tests (Walds test) and the exact test in the Normal and Student-t case.
• For other exact tests, the question will walk you through it.
• Understand how to compute p-value and that the α level controls the probability of type I
error.

Format of Exam
The exam is 3 hours long and you will only need a pen and a calculator (to make you feel good).
There will be four questions and it will basically be

• Question 1: Discrete random variables, probability, set theory, reading comprehension type.
• Question 2: Continuous r.v and Bivariate r.vs
• Question 3: Limit theorem and inference

• Question 4: Statistical inference (Or anything goes!)

Formula sheet will be given.

R-studio
• Coding in R will not be tested
• Able to read existing R-code to find p-value, quantile values.

You might also like