What is Model Explainability and how can it give you more insight into Machine Learning predictions?
Survival analysis for employee churn – Part 1: Introduction
Are employees who drive to work more likely to leave compared to those who cycle? In this 5 minute video on survival analysis for employee churn, Data Scientist Thomas Stainer explains how using survival analysis can help answer questions like the one above, based on a rather contrived data set.
Survival analysis is prevalent in medicine but it’s also widely applicable to areas outside of medicine. In marketing, for example, it’s used to understand customer retention. In fact, anything that is concerned with the time until an event, survival analysis can be applied.
Thomas examines how we can use survival analysis to gain insights into employee churn. If you have the right data about employees and their attrition, you can start testing certain hypotheses.
The key is to use your data to get actionable insights on your workforce and then to act on those insights and repeat: TEST – LEARN – EXPERIMENT – IMPROVE.
Want to read more? Go to the full article.
Censoring is an important feature in Survival Analysis that helps you take into account missing data.