Normality Transformation
1. Load Your Data
Statistical Transformation
Select columns and apply statistical transformations
Transform Multiple Columns
Select columns and their transformation methods. Add up to 5 transformations.
1/5
Learn More
Understanding Statistical Transformations
Purpose
Transform skewed data to approximate normality, beneficial for:
- Meeting statistical test assumptions
- Improving model performance
- Making patterns in data more interpretable
Common Transformations
Log Transformation
Best for:
- Right-skewed data (long right tail)
- Values spanning several orders of magnitude
- When relative changes are more important than absolute changes
Limitations:
- Cannot handle zero or negative values without adjustment
- May over-transform highly skewed data
Square Root Transformation
Best for:
- Moderate right skewness
- Count data
- When variance is proportional to the mean
Limitations:
- Cannot handle negative values
- May not be strong enough for severe skewness
Box-Cox Transformation
Best for:
- Finding optimal transformation power
- When other transformations don't provide satisfactory results
- When you need flexibility in transformation strength
Key Lambda (λ) Values:
- λ = 1: No transformation
- λ = 0: Natural log
- λ = 0.5: Square root
- λ = -1: Reciprocal
Evaluating Transformation Success
Check these metrics:
- Skewness: Should be closer to 0 (symmetrical)
- Kurtosis: Should be closer to 3 (normal distribution)
- Visual assessment: Q-Q plots or histograms should appear more normal
- Statistical tests: Shapiro-Wilk or Anderson-Darling test results
Can't find what you need?
If you couldn't find what you're looking for, we'd love to hear from you!
Let us know what you need