Survival analysis predicts how much time passes before an event occurs while accounting for observations whose events have not yet occurred. section 9.1.1
It overlaps with regression but also includes powerful nonregression methods. Survival curves estimate the probability that a sample remains event-free at each time interval. section 9.1.1
Typical questions include:
- When will a factory machine fail? section 9.1.1
- How long will a customer remain subscribed? section 9.1.1
- What factors contribute to users abandoning a paid subscription sign-up process? section 9.1.1
- How does survival differ between treatment and control groups? section 9.1.1
Kaplan–Meier survival curve
A decreasing step curve shows survival probability dropping after successive time intervals. A more pronounced control-group decline suggests that treatment is associated with prolonged survival probability. section 9.1.1
Table 9.1 Classes of models, their purposes, and examples
|
Class
|
Purpose
|
Example
|
|---|---|---|
|
Regression
|
Predict outcomes represented with continuous data
|
Predicting student test scores
|
|
Classification
|
Predict outcomes represented with categorical data
|
Predict which students will graduate on time
|
|
Clustering
|
Identify underlying patterns and natural “groups” in the data
|
Find clusters of student performance across classes
|
|
Dimension reduction
|
Identify shared variance among predictors to simplify model input
|
Find 2 underlying components to represent scores in 12 classes
|
|
Survival analysis
|
Predict the time until an event occurs
|
Predict the time to withdraw from an intensive graduate curriculum
|
Compare survival analysis with regression, classification, clustering, and dimension reduction by the problem each model class solves. section 9.1.1
