Compact Guide to Classical Inference
Preface
1
What is classical inference?
… and why should I read this book?
2
Data and variables
2.1
Data frames
2.2
Tabulations
2.3
Quantitative and categorical variables
2.4
Response and explanatory variables
3
Measuring variation
3.1
Variance of a numerical variable
3.2
Variance of a categorical variable?
4
Modeling variation
4.1
Statistical models
4.2
Quantitative response variables
4.3
Proportions and indicator variables
4.4
A taxonomy of simple models
5
Model values
5.1
Model fitting: A contest between candidate models
5.2
Variance of model values
6
Degrees of flexibility
6.1
One degree of flexibility
6.2
Multiple degrees of flexibility
6.3
Covariates
6.4
Flexibility, literally
6.5
Degrees of flexibility and freedom
7
Effect size
7.1
With respect to …
7.2
Slopes and differences
7.3
Risk
7.4
Simple changes in input
7.5
Reading effect size from a graph
8
F and R
8.1
The F statistic
8.2
What’s the meaning of F?
8.3
R-squared
8.4
F in statistics books
8.5
Another explanation of F
9
Confidence intervals
9.1
Calculating confidence intervals on effect size
9.2
4 and 95%
9.3
4 and
\(n\)
9.4
Confidence versus prediction intervals
9.5
For the conventionally trained reader …
10
So-called “statistical significance”
10.1
Calculating a p-value
10.2
History and criticism
10.3
Appendix: When
\(df \geq 2\)
11
Simple means and proportions
11.1
No flexibility: df = 0
12
Comparing models
13
Outside of the normal
14
Remember, inference isn’t everything
14.1
Techniques when “degrees of freedom” are unknown, as with machine learning methods.
Published with bookdown
A Compact Guide to Classical Inference
Chapter 12
Comparing models
Looking at
\(\Delta F\)
.