Survival Analysis

Introduction to Survival Analysis

Survival Analysis

Workshop Overview:

The series of videos focuses on survival analysis, a statistical approach to analyze time-to-event data.

Key Topics Covered:

1. Basic Concepts:
   • Survival analysis examines the distribution of times for events to occur.
   • The focus is on estimating survival and hazard functions.
2. Workshop Materials:
   • Resources are available on GitHub.
   • The workshop uses the book “Applied Survival Analysis Using R” by Dirk F. Moore.
3. Example Datasets:
   • Telco Churn Data: Customer usage information and subscription churn.
   • Prostate Survival Data: Cancer survival study.
   • Smoker Data: Smoking relapse study with different treatments.
4. Basic Principles:
   • Data Censoring and Truncation: Handling incomplete observations (right-censored, left-censored data).
   • Hazard and Survival Functions: Defining survival distribution, survival function (\(S(t)\)), and hazard function (\(\lambda(t)\)).
   • Cumulative Functions: Using cumulative hazard function (\(\Lambda(t)\)) for survival estimation.
   • Mean and Median Survival: Calculating expected and median survival times.
5. Example Distributions:
   • Different hazard scenarios (constant, early, late hazard) are explored using R plots.
6. Estimations of the Survival Functions:
   • Kaplan-Meier Estimator: Standard method to estimate survival function.
   • Practical application using R’s survfit() function and visualizations with the survminer package.

Practical Implementation:

The workshop provides practical examples using R code to demonstrate survival analysis concepts, data handling, and visualizations.