This Exam P sample reference tests Joint Distributions. Only outcomes with Y=1 contribute to E[Y/X]. Each such joint mass is 1/18 for X=1,2,3, so the expectation is (1+1/2+1/3)/18=11/108 and choice B is correct.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 3/38 equals E[Y]/E[X]: here E[Y]=1/6 and E[X]=19/9. In general, the expectation of a ratio is not the ratio of expectations.
CThis is three times the correct value, as if each contributing joint mass were 1/6 rather than the specified 1/18.
DThis averages the two possible values of Y as though they were equally likely and ignores both X in the denominator and the joint probabilities.
EThis is E[1/X | Y=1]=(1+1/2+1/3)/3=11/18. It conditions away the probability P(Y=1)=1/6 that must remain in the unconditional expectation.
Original practice · fully worked
Original variant: expected flag score under a random crew size
An inspection station assigns one inspector with probability 0.40 and three inspectors with probability 0.60. Conditional on the assigned crew size X, each inspector independently raises a flag with probability 0.50; Y is the total number of flags. The station records the score S=Y/(X+1). Calculate E[S].
A 0.2500
B 0.3125
C 0.3250
D 0.3750
E 0.5000
Variant answer in brief
Given X=x, the binomial flag count has mean 0.5x, so E[S|X=x]=0.5x/(x+1). The one- and three-inspector conditional scores are 0.25 and 0.375; weighting them by 0.40 and 0.60 gives 0.325, choice C.
Setup
Setup
Condition on the assigned crew size. The flag count is binomial within each crew.
Y∣X=x∼Binomial(x,0.50)
E[Y∣X=x]=0.50x
Model
Model
For a fixed crew size, the score denominator is constant, so conditional expectation may be applied to the numerator.
E[S∣X=x]=x+1E[Y∣X=x]=x+10.50x
E[S∣X=1]=0.25,E[S∣X=3]=0.375
Compute
Compute
Average the two conditional scores using the crew-size probabilities.
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.