This Exam P sample reference tests Joint Cumulative Distribution Functions. The factorized joint CDF gives marginal masses P(X=3)=0.5³(1-0.5)=0.0625 and P(Y=3)=0.2³(1-0.2)=0.0064. Their product is 0.00040, choice B.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.00004 is P(X=4,Y=4)=(0.03125)(0.00128). It shifts both CDF differences one count too far.
CUsing F(3,1)=0.9000 as the added corner in the four-term difference gives 0.9360-0.8736-0.9300+0.9000=0.0324, near 0.03244. The correct lower-left corner is F(2,2).
DThe value 0.06800 is F(3,3)-F(2,2)=0.9360-0.8680. A diagonal difference includes two boundary strips, not just the single cell.
EThe value 0.12440 is [F(3,3)-F(2,3)]+[F(3,2)-F(2,2)]=0.0624+0.0620. It adds two nested X=3 strips instead of differencing them to isolate Y=3.
Original practice · fully worked
Original variant: covariance recovered from a binary joint CDF
Binary alarm indicators U and V take values 0 or 1. Their joint CDF G satisfies G(0,1)=0.55, G(1,0)=0.40, and G(0,0)=0.25. Calculate Cov(U,V).
A 0.0300
B 0.2500
C 0.3000
D 0.4500
E 0.6000
Variant answer in brief
The CDF values give P(U=1)=0.45 and P(V=1)=0.60. Inclusion-exclusion gives P(U=1,V=1)=0.30, so Cov(U,V)=0.30-(0.45)(0.60)=0.03, choice A.
Setup
Setup
Because the indicators are binary, two boundary CDF values are marginal zero probabilities.
Pr(U=0)=G(0,1)=0.55,Pr(V=0)=G(1,0)=0.40
E[U]=0.45,E[V]=0.60
Model
Model
Use the lower-left CDF corner to recover the probability that both indicators equal one.
Pr(U=1,V=1)=1−Pr(U=0)−Pr(V=0)+Pr(U=0,V=0)
Compute
Compute
Evaluate the joint one-cell and subtract the product of the marginal means.
E[UV]=1−0.55−0.40+0.25=0.30
Cov(U,V)=E[UV]−E[U]E[V]
Cov(U,V)=0.30−(0.45)(0.60)=0.03
Answer
Answer
The covariance of the two alarm indicators is 0.03.
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.