This Exam P sample reference tests Mixture Distributions. This is a mixture probability over four latent count states. Weighting each conditional zero-visit probability gives 0.389948, so the probability of one or more visits is 0.610052 and choice C.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.18 is only the probability of the highest latent illness state and does not aggregate visits across the other states.
BThe value 0.39 is the probability of zero visits, 0.389948 rounded, so it is the complement of the requested event.
DThe value 0.72 is Pr(X>0). It assumes every positive illness count guarantees a visit, ignoring the positive Poisson probability of zero.
EOmitting the structural zero-state term gives 1-[0.12 exp(−1)+0.42 exp(−2)+0.18 exp(−3)]=0.890052, near 0.89.
Original practice · fully worked
Original variant: passes from a random scanner panel
A control panel activates 1, 2, or 4 scanners with probabilities 0.25, 0.50, and 0.25, respectively. Conditional on the number activated, scanners pass a diagnostic independently with probability 0.80. Calculate the probability that at least two scanners pass.
A 0.2432
B 0.3200
C 0.5632
D 0.6400
E 0.9728
Variant answer in brief
The one-scanner state contributes zero. The two- and four-scanner states contribute 0.50(0.64)=0.32 and 0.25(0.9728)=0.2432, totaling 0.5632 and choice C.
Setup
Setup
Let N be the active-scanner count and K the number that pass. Conditional on N, K is binomial.
K∣N=n∼Binomial(n,0.80)
Model
Model
Evaluate the target under each possible panel size.
Pr(K≥2∣N=1)=0
Pr(K≥2∣N=2)=(0.80)2=0.64
Pr(K≥2∣N=4)=1−(0.20)4−4(0.80)(0.20)3=0.9728
Compute
Compute
Weight the three conditional probabilities by the panel-size probabilities.
Pr(K≥2)=0.25(0)+0.50(0.64)+0.25(0.9728)
Pr(K≥2)=0.3200+0.2432=0.5632
Answer
Answer
The probability that at least two scanners pass is 0.5632.
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.