Sums of independent random variables

Sums of independent random variables

Spec SD8 · AQA A-Level Further Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • E(X+Y)=E(X)+E(Y)E(X+Y)=E(X)+E(Y) always, with or without independence.
  • For independent variables, Var(X+Y)=Var(X)+Var(Y)\mathrm{Var}(X+Y)=\mathrm{Var}(X)+\mathrm{Var}(Y).
  • Add variances, never standard deviations.
  • Two separate observations: Var(X1+X2)=2σ2\mathrm{Var}(X_1+X_2)=2\sigma^2; one value doubled: Var(2X)=4σ2\mathrm{Var}(2X)=4\sigma^2.

The mean of a sum

The means always add, for any two random variables.

For any two random variables, discrete or continuous:

E(X+Y)=E(X)+E(Y).E(X+Y)=E(X)+E(Y).

This is true whether or not XX and YY are independent, and it works if one is discrete and the other continuous.

If E(X)=12E(X)=12 and E(Y)=7E(Y)=7 then E(X+Y)=19E(X+Y)=19. This is the expectation of a sum.

05101520E(X)E(Y)E(X + Y)Mean
E(X + Y) = E(X) + E(Y) = 12 + 7 = 19
  • E(X+Y)E(X+Y)E(X)+E(Y)E(X)+E(Y)
  • Needs independence?no

E(X)=5E(X)=5 and E(Y)=8E(Y)=8, but XX and YY are not independent. What is E(X+Y)E(X+Y)?

The variance of a sum

For independent variables the variances add; take the square root last to get a standard deviation.

If XX and YY are independent:

Var(X+Y)=Var(X)+Var(Y).\mathrm{Var}(X+Y)=\mathrm{Var}(X)+\mathrm{Var}(Y).

With Var(X)=9\mathrm{Var}(X)=9 and Var(Y)=16\mathrm{Var}(Y)=16, Var(X+Y)=25\mathrm{Var}(X+Y)=25 and the standard deviation is 55, not 3+4=73+4=7.

If a question gives pdfs, find EE and Var\mathrm{Var} for each variable separately (using Var=E(X2)−[E(X)]2\mathrm{Var}=E(X^2)-[E(X)]^2), then combine.

0510152025XYX + YValue
  • Variance
  • Standard deviation
Variances add (9 + 16 = 25) but standard deviations do not (3 + 4 ≠ 5)
  • Var(X+Y)\mathrm{Var}(X+Y)Var(X)+Var(Y)\mathrm{Var}(X)+\mathrm{Var}(Y)
  • Standard deviationsquare root of the total variance

Worked example

WW has pdf 16\frac16 on [4,10][4,10] and BB has pdf b8\frac{b}{8} on [0,4][0,4], independent. Find E(T)E(T) and Var(T)\mathrm{Var}(T) for T=W+BT=W+B.

Independent XX and YY have Var(X)=25\mathrm{Var}(X)=25 and Var(Y)=144\mathrm{Var}(Y)=144. What is the standard deviation of X+YX+Y?

Independence matters

Only the variance result needs independence; state it in context when you use it.

The variance result is only valid when XX and YY are independent. State the assumption in context: for example, the preparation time for a dish does not affect its cooking time.

Independence is not needed for the mean.

0246810121416IndependentY = XY = −XVar(X + Y)
Var(X) = Var(Y) = 4: only when X and Y are independent does Var(X + Y) equal Var(X) + Var(Y) = 8

Mean of a sum

Result:
E(X+Y)=E(X)+E(Y)E(X+Y)=E(X)+E(Y)
Needs independence?:
No

Variance of a sum

Result:
Var(X)+Var(Y)\mathrm{Var}(X)+\mathrm{Var}(Y)
Needs independence?:
Yes

Which result needs X and Y to be independent?

Two observations or doubling

No: two separate observations give 2σ22\sigma^2, but doubling one value gives 4σ24\sigma^2.

Let X1X_1 and X2X_2 be independent observations from the same distribution, with mean μ\mu and variance σ2\sigma^2.

E(X1+X2)=2μ,Var(X1+X2)=2σ2.E(X_1+X_2)=2\mu,\qquad \mathrm{Var}(X_1+X_2)=2\sigma^2.

Doubling a single observation is different: Var(2X)=4σ2\mathrm{Var}(2X)=4\sigma^2, because the same value is used twice, so the variation is not averaged out.

01234X₁ + X₂2XVariance / σ²
Two independent observations have variance 2σ²; doubling one observation has variance 4σ²

X1+X2X_1+X_2 (two values)

Mean:
2μ2\mu
Variance:
2σ22\sigma^2

2X2X (one value doubled)

Mean:
2μ2\mu
Variance:
4σ24\sigma^2

Two independent observations come from a distribution with σ2=5\sigma^2=5. What is Var(X1+X2)\mathrm{Var}(X_1+X_2)?

Try an exam question

Independent random variables XX and YY have E(X)=12E(X)=12, Var(X)=9\mathrm{Var}(X)=9, E(Y)=7E(Y)=7 and Var(Y)=16\mathrm{Var}(Y)=16. Find E(X+Y)E(X+Y) and the standard deviation of X+YX+Y.

[3 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Find E(X+Y)E(X+Y).

1 mark, marked against the mark scheme

Revision notes