ms4ds
1
Outline
2
Introduction
3
Ethical data science
3.1
Motivating example: fair algorithms
3.1.1
A brief historical interlude
3.2
Philosophy of ethics
3.2.1
Meta-ethics
3.2.2
Normative ethics
3.3
Applied ethics
3.3.1
Belmont-Menlo principles
3.3.2
The five Cs framework
3.3.3
Ethical Guidelines for Statistical Practice
3.4
Causality and ethics
3.4.1
Causal models
3.4.2
Moral luck
3.5
Additional reading
4
Foundations of inference
4.1
Philosophy of statistics
4.1.1
Additional reading
4.2
Hypothesis testing background
4.2.1
Hypothesis formulation: a neglected topic in data science education?
4.3
Causality
5
Large scale testing
5.1
Multiple testing
5.1.1
Family-wise error rate
5.1.2
Special cases: groups and hierarchy
5.2
Selective inference
5.2.1
False discovery rate
5.2.2
Special cases: groups and hierarchy
5.3
Empirical Bayes approaches
6
Causal inference
6.1
A variety of treatment effects
6.2
Directed acyclic graphs and structural equation models
6.2.1
Common problems?
7
Foundations of regression
7.1
Philosophy of regression models
7.2
Linear regression
7.3
Generalized linear models and exponential families
7.4
Inference in regression models
7.5
Model diagnostics
8
Machine learning and high dimensional regression
8.1
Lasso and penalized regression
8.2
Generalized additive models
8.3
Trees, forests, boosting
9
Selective inference
9.1
Conditional approach
9.2
Debiasing approaches
10
Machine learning for causal inference
10.1
Double machine learning
10.2
Weight-based approaches
References
Published with bookdown
Modern Statistics 4 Data Science
Section 7
Foundations of regression
7.1
Philosophy of regression models
Why assume errors are Gaussian?
False hope of causality
7.2
Linear regression
7.3
Generalized linear models and exponential families
7.4
Inference in regression models
7.5
Model diagnostics