Properties of Expectation and Variance
What you'll learn: How expectation and variance behave under mathematical transformations, making complex probability calculations much simpler.
You've learned to calculate expected value and variance for discrete random variables. Now you'll discover powerful shortcuts that let you work with transformed or combined variables without recalculating everything from scratch.
Linearity of Expectation
The most useful property: expectation is linear. This means:
-
Adding a constant: If you add a constant c to a random variable X, the expected value increases by exactly c:
E(X + c) = E(X) + c -
Multiplying by a constant: If you multiply X by constant a, the expectation scales proportionally:
E(aX) = a · E(X) -
Combined:
E(aX + b) = a · E(X) + b
Example: If average test scores have E(X) = 70, and you curve grades by multiplying by 1.1 and adding 5 points, the new average is 1.1(70) + 5 = 82.
Variance Under Linear Transformations
Variance behaves differently because it measures spread, not location:
-
Adding a constant doesn't change spread:
Var(X + c) = Var(X)— shifting all values doesn't affect variability -
Multiplying scales variance by the square:
Var(aX) = a² · Var(X)— doubling values quadruples variance -
Combined:
Var(aX + b) = a² · Var(X)
Analogy: If you convert temperatures from Celsius to Fahrenheit (multiply by 9/5, add 32), the average shifts by the full formula, but the variance only depends on the 9/5 multiplier—and gets squared.
Why These Matter
These properties let you calculate statistics for transformed distributions instantly, without redoing entire probability tables. They're foundational for combining multiple random variables later.
Key Takeaway: Expectation shifts linearly with transformations (E(aX + b) = aE(X) + b), while variance only responds to scaling—squared (Var(aX + b) = a²Var(X)).