Uncertainty and models
Survival / Kaplan-Meier curve
Estimates the probability of remaining event-free beyond each time while accounting for censoring.
- Plot
- Time to event
- Comparison
What it answers
What fraction remains event-free over time?
- Shape
- Descending steps and censor marks
- What it shows
- Event timing and group survival differences.
Example
The question
Do customers on annual plans stay longer than monthly ones?
Share of customers still subscribed by months since joining (Kaplan-Meier)
Illustrative data. It belongs to no company.
When to use it
Use it when
Time-to-event data with valid censoring assumptions.
Avoid it when
Treating censored follow-up as a known event-free lifetime.
How to read it
Follow descending steps and distinguish event drops from censor marks.
Note
Event and censoring indicators are essential.
The data you need
- Layout
- Long table
- Fields
- SubjectID: text; Duration: nonnegative number; Event: 0/1
- One row is
- One subject follow-up, ending in an event or censoring. What is the grain?
- Rules
- Define Event=1 and censoring=0, a common time origin and time units. Use appropriate methods for truncation or competing risks.
- Optional
- Group,EntryTime. A precomputed curve needs Time,Survival or CumulativeHazard and optional bounds.
- Preparation
- Define time origin and event/censor indicators; calculate risk sets and a Kaplan-Meier estimate under appropriate assumptions.
- Units
- Survival probability uses successive event-risk sets; time uses declared follow-up units. Censoring changes risk sets, not event counts.
Worked example
Three subjects: event 12, censored 18, event 25. Kaplan-Meier survival drops to 2/3 at 12, stays at 18, then drops to 0 at 25 with one subject at risk.
| SubjectID | Duration | Event |
|---|---|---|
| S1 | 12 | 1 |
| S2 | 18 | 0 |
| S3 | 25 | 1 |