Linear Transformations of Normal Variables
What you'll learn: How shifting or scaling a normal distribution changes its mean and standard deviation in predictable ways.
The Core Idea
One of the most elegant properties of the normal distribution is that it stays normal even when you transform it. If you start with a normally distributed variable and either add/subtract a constant or multiply/divide by a constant, the result is still normally distributed—just with updated parameters.
Two Simple Rules
Rule 1: Adding or subtracting a constant
If X ~ N(μ, σ) and you create Y = X + c, then:
- The mean shifts by
c: new mean =μ + c - The standard deviation stays the same:
σ
Example: If test scores are N(70, 10) and you add 5 bonus points to everyone, the new distribution is N(75, 10). Everyone's score shifts up, but the spread doesn't change.
Rule 2: Multiplying or dividing by a constant
If X ~ N(μ, σ) and you create Y = aX, then:
- The mean scales by
a: new mean =aμ - The standard deviation also scales by
|a|: new std =|a|σ
Example: If heights in inches are N(68, 3) and you convert to centimeters by multiplying by 2.54, you get N(172.72, 7.62). Both the center and spread scale proportionally.
Combined Transformations
You can combine these: Y = aX + b gives you N(aμ + b, |a|σ).
Key Takeaway: Linear transformations of normal variables produce new normal distributions with predictable parameters: adding shifts the mean, multiplying scales both the mean and standard deviation, and the normality is preserved.