Calculating Information Value
So far you have combined data from multiple sources and created new variables to derive insights from data. Do you think all these variables can explain turnover?
Information Value (IV) helps in measuring and ranking the variables on the basis of the predictive power of each variable. You can use Information Value (IV) to drop the variables which have very low predictive power.
This is a part of the course
“HR Analytics: Predicting Employee Churn in R”
Exercise instructions
- Load the
Informationpackage. - Use the
emp_finaldataset from the previous exercise to find information value of all the variables in the dataset. - Print Information Value (IV) of each variable.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Load Information package
___
# Compute Information Value
IV <- create_infotables(data = ___, y = ___)
# Print Information Value
___This exercise is part of the course
HR Analytics: Predicting Employee Churn in R
Predict employee turnover and design retention strategies.
In this chapter, you will create new variables from existing data to explain employee turnover. You will analyze compensation data and create compa-ratio to measure pay equity of all employees. To identify the most important variables influencing turnover, you will use the concept of Information Value (IV).
Exercise 1: Feature engineeringExercise 2: Derive age differenceExercise 3: Derive job hop indexExercise 4: Derive employee tenureExercise 5: CompensationExercise 6: Exploring compensationExercise 7: Pay GapExercise 8: Deriving Compa-ratioExercise 9: Deriving Compa-levelExercise 10: Information valueExercise 11: Calculating Information ValueExercise 12: Which variables are important?What is DataCamp?
Learn the data skills you need online at your own pace—from non-coding essentials to data science and machine learning.