DSCI 552: Statistical Inference and Computation I

The statistical and probabilistic foundations of inference, the frequentist paradigm. Concepts first introduced through simulation, followed with an introduction and exploration of asymptotic theory.

Sample lecture:

Click on the Binder launch button below to view a sample lecture in an interactive notebook:

Learning Outcomes

By the end of the course, students are expected to be able to:

1. Describe the frequentist paradigm of statistical inference.
2. Select and compute appropriate estimates of statistical parameters and distributional quantities using bootstrapping and asymptotic methods.
3. Construct and run simulations to evaluate uncertainty of estimates and predictions in hypothetical situations.
4. Evaluate uncertainty (e.g., confidence intervals) of estimates and predictions from data using bootstrapping and asymptotic results.
5. Explain the difference between prediction and estimation error in terms of irreducible and reducible error.
6. Carry out basic hypothesis tests in a responsible manner, and responsibly apply them in appropriate situations.
7. Explain the fundamental ideas and terminology of hypothesis testing.
8. Perform estimation using Maximum Likelihood.

Assessments

This is an assignment-based course. You'll be evaluated as follows:

Assessment	Weight	Deadline	Location
Lab 1 - Understanding sampling through simulation and bootstrapping	20%	2018-10-20 @ 18:00	Submit to Github
Lab 2 - Confidence Intervals	15%	2018-10-27 @ 18:00	Submit to Github
Lab 3 - Hypothesis Testing	15%	2018-11-10 @ 18:00	Submit to Github
Lab 4 - Asympototic theory and Maximum Likelihood estimation	10%	2018-11-06 @ 18:00	Submit to Github
Quiz 1	20%	2018-10-29 @ 14:00	On Canvas and written in your lab room
Quiz 2	20%	2018-11-08 @ TBD	On Canvas and written in TBD

Tip: Use the lecture learning objectives as beacons when studying for your quizzes!

Schedule

Lecture	Topic	Readings
1	Understanding sampling through simulation	Sampling
2	Bootstrapping	Bootstrapping Comparing bootstrap and sampling distributions
3	Confidence intervals using bootstrapping	Confidence Intervals
4	Introduction to Hypothesis testing	Hypothesis Testing
5	There is only one test	There is only one test! There is still only one test
6	An introduction to Asympototic theory
7	Using Asympototic theory for confidence intervals and hypothesis tests
8	Maximum Likelihood Estimation

Textbooks

1. Modern Dive: An Introduction to Statistical and Data Sciences by R by Chester Ismay and Albert Y. Kim
2. OpenIntro Statistics by David M Diez, Christopher D Barr and Mine Cetinkaya-Rundel