This Exam P sample reference tests Bayes' Theorem. The zero-count relationship determines the first class's Poisson mean as 3-ln(3)=1.901388. Combining the two exact-count likelihoods with prior weights 0.40 and 0.60 gives posterior probability 0.445490, so choice D is correct.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThis is the joint probability Pr(S and X=2), approximately 0.108. It is the Bayes numerator before normalization.
BThis is the likelihood Pr(X=2 given S), approximately 0.270. It does not incorporate either prior class weight.
CThis is the prior probability of class S. Observing the count changes that probability because the two Poisson likelihoods differ.
EThis is the posterior complement, the probability of the other class after the count is observed.
Original practice · fully worked
Original variant: source allocation after a combined count
Two independent particle counters register Poisson counts with means 2 and 3 during the same interval. The combined display reports exactly four particles. Conditional on that total, calculate the probability that the first counter registered at least two particles.
A 0.1296
B 0.3456
C 0.4752
D 0.5248
E 0.8704
Variant answer in brief
Given the combined total, the first source count is binomial with four trials and success probability 2/5. Its probability of at least two is 1-0.1296-0.3456=0.5248, selecting choice D.
Setup
Setup
Let X and Y be the independent source counts and T their observed sum.
X∼Poisson(2),Y∼Poisson(3)
T=X+Y=4
Model
Model
Independent Poisson sources split a fixed combined total in proportion to their means.
X∣(T=4)∼Binomial(4,2+32)
Compute
Compute
Use the complement of zero or one particle from the first counter.
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.