Linear Transformations of Random Variables (AI HL)
If , then and , so you can find the mean and variance of without rebuilding its distribution. The same ideas extend to combining independent variables, where means add or subtract but variances always add. Worked examples show both rules in use. It's part of the broader Probability & Distributions topic.
22 questions on this sub-topic.
Syllabus
| Code | Syllabus content |
|---|---|
| SL4.7 | Concept of discrete random variables and their probability distributions, given as a table or a formula. Expected value for discrete data, and its applications - including that indicates a fair game where is a player's gain. |
Official IB source: Mathematics: applications and interpretation subject brief (PDF, ibo.org).
The two rules
This builds on IB syllabus reference SL4.7 - the concept of discrete random variables, their probability distributions, and expected value . Neither rule below is in the formula booklet, but both follow directly from the definitions of and .
Expected value of
Scale and shift the mean exactly as you scale and shift itself.
Variance of
Adding a constant doesn't spread the data out any further, so it drops out entirely; gets squared.
Need the full syllabus wording and formula-booklet reference table? See Probability & Distributions. For calculator routines, see the parent topic's GDC guidance.
Key terms
What does a linear transformation do to a random variable?
A linear transformation multiplies every value of by and then adds . The probabilities stay the same; only the values move.
e.g. If takes the values , then takes the values with the same probabilities.
Why does adding a constant not change the variance?
Adding shifts every value by the same amount, so the spread about the mean is unchanged. Variance measures spread, not position.
e.g. If , then .
How does the standard deviation change under a linear transformation?
The variance is multiplied by , so the standard deviation is multiplied by . A standard deviation is never negative.
e.g. If has standard deviation , then has standard deviation .
What are independent random variables?
Two random variables are independent when the value taken by one does not affect the probabilities of the other. For independent and , .
e.g. If independent and have and , then .
Worked examples
A discrete random variable has and . Let .
(a) Find .
(b) Find .
Worked solution
(a) M1
A1
(b) M1
A1
A stall sells items. Revenue has and . Costs (independent) have and Profit
(a) Find
(b)(i) Find
(b)(ii) Find the standard deviation of profit.
Worked solution
(a) Subtract the means: M1
A1
(b)(i) Variances add: M1
A1
(b)(ii) The standard deviation is A1On the GDC. (a) Find E(P). (b)(i) Find . (b)(ii) Find the standard deviation of profit
has .
Find
Worked solution
M1
A1
Common mistakes
- Forgetting to square in the variance rule. , not - a negative still gives a positive multiplier once it's squared.
- Carrying the constant into the variance. A constant shift changes where the data is centred but not how spread out it is, so simply disappears from - writing is a very common slip.
- Subtracting variances when combining independent variables. Even for , the variances of two independent variables always add: , never .
Using your GDC
Every step below is a real button sequence, not a vague "use your calculator" hint - covering the TI-84 Plus, TI-Nspire, and Casio fx-9860/fx-CG50. Pick your model to filter down to just the steps that apply to you.
The reverse of a normal probability: given a percentage, find the cut-off value (e.g. the mark for the top 10%).
- Work out the area to the LEFT of the value you want.
- 2nd → VARS (DISTR) → invNorm(area, μ, σ). Newer OS lets you pick the tail.TI-84
- menu → Probability → Distributions → Inverse Normal; enter the area, μ and σ.Nspire
- Main menu → Statistics → DIST → NORM → InvN; set the tail and enter area, σ, μ.Casio
Tip: invNorm needs the area to the LEFT. For "top 10%", use area = 0.90; for "bottom 25%", use area = 0.25.
Given a probability and one parameter, work back to the missing or - a standard exam twist.
- Turn the probability into a -value using the inverse normal with , .
- invNorm(area-to-left, 0, 1) gives ; then solve for the unknown.TI-84
- Use Inverse Normal with , to get , then solve the standardising equation for or .Nspire
- DIST → NORM → InvN with , gives ; substitute into .Casio
- If two probabilities are given, form two equations and solve them simultaneously for and .
Tip: Always sketch the curve first so you can tell which side of the mean each value should fall on.
See the full GDC guide for more calculator models and topics.
Quick questions
What is E(aX + b) in terms of E(X)?
- scale and shift the mean the same way you scale and shift .
What is Var(aX + b) in terms of Var(X)?
- the constant has no effect on variance, and is squared.
Why is the variance multiplied by a squared?
Variance is based on squared distances from the mean. Scaling by scales each distance by and so each squared distance by , which is also why a negative never gives a negative variance.
How do I find the mean of X - Y?
Subtract the means: . If and , then .
Is Var(2X) the same as Var(X + X)?
Yes, because , so both equal . Adding two independent copies of is different and gives , since the two values can differ.
Ready to practise properly?
22 linear-transformation questions, marked instantly like the real exam.