What is survival analysis?

Survival Analysis in Python

Shae Wang

Senior Data Scientist

What is survival analysis?

Medical research survival analysis books.

Survival Analysis in Python

What is survival analysis?

A branch of statistics focused on analyzing time to an event.

Examples

  • Time until equipment failure
  • Time until loan default
  • Time until free-trial users convert to subscribers
  • Time until a tree's first fruit
Survival Analysis in Python

What is survival analysis?

A branch of statistics focused on analyzing time to an event.

  • Event: Equipment failure, loan default, disease recurrence, user conversion, or other experience of interest
  • Survival: The event of interest does not occur
  • Survival duration: Time until the event of interest occurs (or the end of our observations)
Survival Analysis in Python

Time-to-event data

time to event data diagram

ID Duration Observed
A 2 Yes
B 4 Yes
C 5.2 No
D 2.5 Yes
  • Event: must be clear and unambiguous, cannot partially happen
  • There should be abundant data where the event does occur for survival analysis to be effective.
Survival Analysis in Python

The use cases of survival analysis

  • Calculate the proportion of subjects that will experience an event
  • Estimate the time until event/failure in a population
  • Calculate the rate of event in a population
  • What factors increase or decrease survival?
  • Predict probabilities of event occurrence for subjects
Survival Analysis in Python

Predict battery failure time

Battery failure time cartoon.

Survival Analysis in Python

Battery time-to-event data

Battery ID Duration Dead Brand Truck
1 2.5 yrs No Brand A Long
2 6 yrs Yes Brand B Short
3 5 yrs No Brand B Long
... ... ... ... ...
1000 4.5 yrs Yes Brand A Short
  • Duration: how long the battery has been in use
  • Dead: whether the battery has died
Survival Analysis in Python

Let's practice!

Survival Analysis in Python

Preparing Video For Download...