Computational Statistics

This is main resources site for the Computational Statistics course. This is where all the material, including slides, assignments, and data, are collected.

Schedule

Note on the exercises:

Each Thursday, our exercise session will center around a key topic. Together, we’ll work through selected exercises on this topic. The remaining exercises are for you to solve independently at your own pace.

Week 1

Lecture 1: Introduction

• Slides: Introduction and Functions in R (source)
• Reading: Chapter 6 (AdvR)
• Exercises: A1-A7 (CSWR, Appendix A)
  • Data: infrared.txt

Exercise Session 1

• Main topic: Setting up a proper development environment for R/Rcpp
• In class walkthrough: A1-A7 (CSwR, Appendix A)
• Solutions: Exercise 1 Solutions

Lecture 2: OOP and Density Estimation

• Slides: OOP and Density Estimation (source)
• Reading:
  • Chapter 13 (AdvR)
  • Sections 2-2.4 (CSwR)
• Exercises:
  • A8-A10 (CSwR, Appendix A)
  • 2.1-2.6 (CSwR, Chapter 2)

Week 2

Lecture 3: Measuring and Improving Performance

• Slides: Measuring and Improving Performance (source)
• Reading: Chapter 23 and Chapter 24 (Advanced R)
• Exercises: 1-3 (Section 24.4.3, AdvR)

Exercise Session 2

• Main topic: Using AI properly, memory usage and benchmarking, dependency injection, reverse engineering
• In class walkthrough: Exercises 2.1-2.6 (CSwR, Chapter 2) + 24.4.3.2 + 24.4.3.3 (AdvR)
• Solutions: Exercise 2 Solutions
• Mini-slides: Tips for using AI

Lecture 4: Parallelization and Scatterplot Smoothing

• Slides: Parallelization and Scatterplot Smoothing (source)
• Reading:
  • Chapter 3, but skip 3.3 (CSwR)
  • Chapter 4, but skip 4.2-4.3 (CSwR)
• Exercises: 3.4 (Chapter 3, CSWR)

Week 3

Lecture 5: Random Number Generation and Rejection Sampling

• Slides: Random Number Generation and Rejection Sampling (source)
• Reading:
  • Chapter 5, but skip 5.1 (CSwR)
  • Chapter 6 (CSwR)
• Exercises:
  • Exercises 5.1-5.2 in Chapter 5 (CSwR)

Exercise Session 3

• Main topic: Premature optimization
• In class walkthrough: Exercises 3.4 (Chapter 3, CSWR) + 5.1-5.2 (Chapter 5, CSWR)
• Solutions: Exercise 3 Solutions

Lecture 6: Monte Carlo Methods and Importance Sampling

• Slides: Monte Carlo Methods and Importance Sampling (source)
• Reading: Chapter 7 (CSwR)

Presentations of Assignment 1

• Instructions: Assignment 1

Week 4

Lecture 7: Optimization

• Slides: Optimization (source)
• Reading:
  • Chapter 8 (CSwR)
  • Chapter 9 (CSwR)

Exercise Session 4

• Main topic: Currying and decorators
• In class walkthrough: Exercises 25.2.6.1, 25.2.6.2, 25.4.5.1, 25.4.5.2(AdvR) + Bonus Example
• Solutions: Exercise 4 Solutions

Presentations of Assignment 2

• Instructions: Assignment 2

Week 5

Lecture 8: Testing and Debugging

This lecture will be mostly a hands-on session. Please bring your laptop. We will work on debugging a piece of code together. You can find the code here: debugging.R.

• Slides: Testing and Debugging (source)
• Reading: Chapter 22 (AdvR)

Exercise Session 5

• Main topic: None - work on your assignments with TA assistance

Lecture 9: Likelihood Optimization and the EM Algorithm

• Slides: Likelihood Optimization and the EM Algorithm (source)
• Reading: Chapter 10 (CSwR)

Week 6

Lecture 10: More on the EM Algorithm

• Slides: The EM Algorithm (source)

Exercise Session 6

• Main topic: None - work on your assignments with TA assistance

Lecture 11: Stochastic Gradient Descent

• Slides: Stochastic Gradient Descent (source)
• Reading: Sections 11-11.1 (CSwR)

Presentations of Assignment 3

• Instructions: Assignment 3

Autumn Holiday Week

Enjoy your break!

Week 7

Lecture 12: Rcpp

Prior to this lecture, you should have Rcpp installed. You can follow the instructions in Rcpp for Everyone to get started.

• Slides: Rcpp (source)
• Reading: Chapter 25 (AdvR)
• Exercises:
  • 1 and 2 from 25.2.6 (AdvR)
  • 1 and 2 from 25.4.5 (AdvR)
  • If you have time over, try the exercises in 25.5.7 (AdvR) too

Exercise Session 7

• Main topic: Branch prediction
• In class walkthrough: TA will go through the topic with a live coding example.
• Solutions: Exercise 7

Lecture 13: Variations on Stochastic Gradient Descent

• Slides: Variations on Stochastic Gradient Descent (source)
• Reading: Sections 11.2-11.3 (CSwR)

Presentations of Assignment 4

• Instructions: Assignment 4

Week 8

Lecture 14: R Packages and Wrap-Up

This lecture will mostly be a hands-on session. Please bring your laptop. We will work on creating an R package together. Please install the devtools package prior to the lecture.

• Slides: R Packages and Wrap-Up (source)

Week 9

Examinations

See the schedule on Absalon.