Homework

All homework should be submitted to Gradescope. You can use the entry code VB756J to enroll in the course.

We have five sets of homework. Each worth 10% of the total score.
You can complete the homework in either R or Python.
Grading is mainly based on
- Correctness of your answers
- Clarity of your code
- Clarity of your writing
- Presentation of your results
You are encouraged to discuss the homework with your classmates. However, you must write your own code and your own answers.
You must submit your homework on Gradescope by the due date. Late homework will be penalized for 5% each day, up to a maximum of 3 days. No homework will be accepted after 3 days.
You must submit your homework in the form of a PDF file. You can consider R Markdown or Jupyter Notebook to write your homework.
You can use ChatGPT or other similar tools to help you. However, you must organize your answers in a clear and logical way. You will be penalized if your answers are not meaningful, especially when the final results are wrong.

Homework 1 [Source File]
Released: Mar 4
Due: Mar 21, 11:59PM CT
This homework mainly covers week 8.
Homework 1 Solution
Corrections: (updated Mar 18)
- In Q2, the total sample size is 50, but since it is an observational study, we cannot force the sample size on each group
- In Q3, we are only interested in ATT (average treatment effect of the treated), not ATE.

Homework 3 [Source File]
Sepsis.csv
Released: Mar 25
Due: Apr 4, 11:59PM CT
This homework mainly covers week 11.
Homework 3 Solution
Corrections: (updated Mar 26)
- In Question 2, you should compare the results between WeightSVM package (vanilla OWL) and DTRlearn2 package (a slightly modified version)
- Also added Question 4