This Exam P sample reference tests Joint Distributions. This is a marginal-variance calculation from a uniform triangular joint support. Counting the admissible x-values gives P(Y=y)=(6-y)/21, from which E[Y]=5/3 and E[Y squared]=5, so Var(Y)=20/9=2.22 and choice B.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 1.67 is the rounded mean E[Y]=5/3. A variance requires the centered second moment, not the first moment.
CThe value 3.33 results from subtracting E[Y] rather than its square: 5-5/3=10/3. The variance identity uses E[Y²]-(E[Y])².
DThe value 5.00 is the raw second moment E[Y²]. It still includes the squared location of the distribution and has not been centered.
EThe value 5.56 results from a misapplied total-variance formula: using the raw second moment 5 as the within-X term and adding Var(E[Y|X])=5/9 gives 50/9. The correct within-X term is E[Var(Y|X)]=5/3.
Original practice · fully worked
Original variant: active checkpoints under a random route class
A dispatch day selects a route class K with probabilities P(K=1)=0.20, P(K=3)=0.50, and P(K=5)=0.30. Given K, each of the K checkpoints is active independently with probability 0.40. Let Y be the number of active checkpoints. Calculate Var(Y).
A 0.3136
B 0.7680
C 1.0816
D 1.2800
E 1.9600
Variant answer in brief
Conditional on K, the active count is binomial. The within-route variance component is 0.7680 and the between-route component is 0.3136, so total variance is 1.0816 and choice C.
Setup
Setup
Compute the first two moments of the random route size.
E[K]=1(0.20)+3(0.50)+5(0.30)=3.20
E[K2]=12(0.20)+32(0.50)+52(0.30)=12.20
Var(K)=12.20−3.202=1.96
Model
Model
Given the route class, the active count is binomial with size K and probability 0.40.
E[Y∣K]=0.40K
Var(Y∣K)=K(0.40)(0.60)=0.24K
Compute
Compute
Add the expected conditional variance and the variance of the conditional mean.
The 3108-page Probability Proof Manual reorganizes 718 verified Exam P solutions by syllabus skill and adds formula proofs, error patterns, and original worked practice.