Exam PA Study Manual
Video lessons
1
How to use this book
2
The exam
3
Prometric Demo
4
You already know what learning is
5
Getting started
5.1
Download the data
5.2
Download ISLR
5.3
New users
6
R programming
6.1
Notebook chunks
6.2
Basic operations
6.3
Lists
6.4
Functions
6.5
Data frames
6.6
Pipes
6.7
The SOA’s code doesn’t use pipes or dplyr, so can I skip learning this?
7
Data manipulation
7.1
Look at the data
7.2
Transform the data
7.3
Exercises
7.4
Answers to exercises
8
Visualization
8.1
Create a plot object (ggplot)
8.2
Add a plot
8.3
Data manipulation chaining
9
Introduction to modeling
9.1
Modeling vocabulary
9.2
Modeling notation
9.3
Ordinary Least Squares (OLS)
9.4
Regression vs. classification
9.5
Regression metrics
9.6
Example
10
Generalized linear Models (GLMs)
10.0.1
Assumptions of OLS
10.0.2
Assumptions of GLMs
10.1
Advantages and disadvantages
10.2
GLMs for regression
10.3
Interpretation of coefficients
10.3.1
Log link
10.4
Other links
11
GLMs for classification
11.1
Binary target
11.2
Count target
11.3
Link functions
11.4
Interpretation of coefficients
11.4.1
Logit
11.4.2
Probit, Cauchit, Cloglog
11.5
Example
12
Classification metrics
12.1
Area Under the ROC Curv (AUC)
12.2
Additional reading
13
Additional GLM topics
13.1
Residuals
13.1.1
Raw residuals
13.1.2
Deviance residuals
13.2
Example
13.3
Log transforms of continuous predictors
13.4
Reference levels
13.5
Interactions
13.6
Offsets
13.7
Tweedie regression
13.8
Combinations of Link Functions and Target Distributions
13.8.1
Gaussian Response with Log Link
13.8.2
Gaussian Response with Inverse Link
13.8.3
Gaussian Response with Identity Link
13.8.4
Gaussian Response with Log Link and Negative Values
13.8.5
Gamma Response with Log Link
13.8.6
Gamma with Inverse Link
14
GLM variable selection
14.1
Stepwise subset selection
14.2
Penalized Linear Models
14.3
Ridge Regression
14.4
Lasso
14.5
Elastic Net
14.6
Advantages and disadvantages
14.7
Example: Ridge Regression
14.8
Example: The Lasso
14.9
References
15
Tree-based models
15.1
Decision Trees
15.1.1
Model form
15.1.2
Advantages and disadvantages
15.2
Ensemble learning
15.2.1
Bagging
15.2.2
Boosting
15.3
Random Forests
15.3.1
Model form
15.3.2
Example
15.3.3
Variable Importance
15.3.4
Partial dependence
15.3.5
Advantages and disadvantages
15.4
Gradient Boosted Trees
15.4.1
Parameters
15.4.2
Example
15.4.3
Advantages and disadvantages
15.5
Exercises
15.5.1
1. RF with
randomForest
15.5.2
2. RF tuning with
caret
15.5.3
3. Tuning a GBM with
caret
15.6
Answers to Exercises
16
Unsupervised Learning
16.1
Principal Componant Analysis (PCA)
16.1.1
Example: PCA on US Arrests
16.1.2
Example: PCA on Cancel Cells
16.2
Clustering
16.2.1
K-Means Clustering
16.3
Hierarchical Clustering
16.3.1
Example: Clustering Cancel Cells
16.3.2
References
17
Practice Exams
18
References
Published with bookdown
Exam PA R Study Manual
17
Practice Exams
Practice exams are available at
ExamPA.net
.