Skip to content

UBC-MDS/DSCI_552_stat-inf-1

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 

Repository files navigation

Welcome to DSCI 552: Statistical Inference and Computation I

This course reviews classical and simulation-based techniques for estimation and hypothesis testing, including inference for means and proportions. We make particular emphasis on case studies and real data sets, as well as reproducible and transparent workflows when writing computer scripts for analysis and reports.

High-Level Goals

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

  • Build a solid foundational understanding of frequentist Statistical Inference (computational and classical!).
  • Become competent using R to perform computation for frequentist Statistical Inference.

Learning Objectives

  • Describe real-world examples of questions that can be answered with the statistical inference methods presented in this course (e.g., estimation, hypothesis testing) and apply inference skills and concepts to answer such questions.
  • Explain what random and representative samples are and how they can influence estimation.
  • Write computer scripts to perform estimation and hypothesis testing via simulation-based inference approaches, as well as by applying results from exact and approximate distributional theory.
  • Interpret and explain results from confidence intervals and hypothesis tests.
  • Compare the application of simulation-based inference approaches with the application of exact and approximate distributional results.
  • Effectively visualize point estimates and different measures of uncertainty (e.g., confidence intervals, standard errors) by writing computer scripts.
  • Discuss the impact of type I & II errors as well as responsible use and reporting of p-values on hypothesis tests.
  • Explain estimator bias and uncertainty, and write a computer script to calculate it.
  • Discuss how an estimator's bias arises (e.g., sample bias, study design), and its implications in statistical inference.
  • Perform all aspects of a statistical analysis (from data consumption to reporting) using reproducible and transparent computer scripts.

Lecture Topics

Lecture Topic References
1 Populations and Sampling Modern Dive: Chapter 7
2 Bootstrapping and its Relationship to the Sampling Distribution Modern Dive: Chapter 8, sections 8 - 8.2 inclusive
3 Confidence Intervals via Bootstrapping Modern Dive: Chapter 8, sections 8.3 - 8.7.1 inclusive
4 Hypothesis Testing via Simulation/Randomization
5 Confidence Intervals Based on the Assumption of Normality or the Central Limit Theorem Modern Dive:
6 Classical Tests Based on Normal and t- Distributions
7 Tests for Multiple Group Comparisons
8 Errors in Inference / There is only one test!

Deliverables

This is an assignment-based course. The following deliverables will determine your course grade:

Assessment Weight
Lab 1 10%
Lab 2 10%
Lab 3 10%
Lab 4 10%
Worksheet 1 1%
Worksheet 2 1%
Worksheet 3 1%
Worksheet 4 1%
Worksheet 5 1%
Worksheet 6 1%
Worksheet 7 1%
Worksheet 8 1%
iClicker 2%
Quiz 1 25%
Quiz 2 25%

Class Schedule & office hours

See calendar.

Textbook

We are using an open source textbook: ModernDive: Statistical Inference via Data Science developed by Chester Ismay and Albert Y. Kim. This book is available as:

Policies

See the general MDS policies.

Attribution

The course is built upon previous years' materials developed by previous instructors.

License

© 2024 Katie Burak, Alexi Rodríguez-Arelis, Le Quan Nguyen, Tiffany Timbers, and Rodolfo Lourenzutti

Software licensed under the MIT License, non-software content licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License. See the license file for more information.

About

No description or website provided.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published