This Exam P sample reference tests Independence. Exactly one loss type is the disjoint union of fire without theft and theft without fire. Independence gives 0.20(0.70)+0.80(0.30)=0.38, so choice B.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.30 is the marginal probability of the second loss type alone. It does not exclude simultaneous occurrence or include the first-only case.
CThe value 0.44 is 0.20+0.30-0.20(0.30), the probability of at least one loss type. It includes the both-types outcome.
DThe value 0.50 is 0.20+0.30. It adds the marginals without removing their intersection.
EThe value 0.56 is 1-0.44, the probability that neither loss type occurs, not exactly one.
Original practice · fully worked
Original variant: covariance with an exclusive-status lamp
Two verification modules return independent pass indicators I and J, with pass probabilities 0.25 and 0.40. A status lamp has indicator Z=1 exactly when one module passes and the other fails. Calculate Cov(I,Z).
A -0.0375
B 0.0000
C 0.0375
D 0.1500
E 0.4500
Variant answer in brief
The lamp probability is 0.25(0.60)+0.75(0.40)=0.45, while IZ=1 only when I=1 and J=0, with probability 0.15. Thus Cov(I,Z)=0.15-(0.25)(0.45)=0.0375, choice C.
Setup
Setup
Compute the probability that the exclusive-status lamp turns on.
E[Z]=Pr(I=J)
=0.25(0.60)+0.75(0.40)=0.45
Model
Model
The product IZ equals one only when module I passes and module J fails.
E[IZ]=Pr(I=1,J=0)=0.25(0.60)=0.15
Compute
Compute
Subtract the product of the two means from the product moment.
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.