This Exam P sample reference tests Law of Total Probability. Weighting each bankruptcy rate by its category share gives an overall bankruptcy probability of 0.098. Service businesses contribute 0.020 of that total, so their conditional share is 0.020/0.098, approximately 0.204. This is choice C.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.020 is the joint probability that a business is both a service business and bankrupt. It is the Bayes numerator, not the posterior probability.
BThe value 0.080 is the bankruptcy rate within the service category. The question reverses the conditioning and asks for the category given bankruptcy.
DThe value 0.250 is the unconditional service-business share. Selection from bankrupt businesses changes the category mix.
EA posterior of 0.308 would require a total bankruptcy probability near 0.065. The weighted contributions from all four categories instead total 0.098.
Original practice · fully worked
Original variant: calibration mode from an interval reading
A controller selects broad calibration mode with probability 0.40 and focused calibration mode with probability 0.60. In broad mode, a reading X is uniform from 0 to 5; in focused mode, X is uniform from 2 to 4. A report shows that X lies between 2.5 and 3.5. Calculate the conditional probability that broad mode was selected.
A 0.080
B 0.200
C 0.211
D 0.300
E 0.500
Variant answer in brief
The reported unit-length interval has likelihood 1/5 in broad mode and 1/2 in focused mode. The prior-weighted contributions are therefore 0.08 and 0.30, so the posterior broad-mode probability is 0.08/0.38=4/19, approximately 0.211 and choice C.
Setup
Setup
Let I denote the reported interval and calculate its likelihood under each uniform mode.
Pr(I∣B)=5−03.5−2.5=0.20
Pr(I∣F)=4−23.5−2.5=0.50
Model
Model
Weight the interval likelihoods by the two mode priors.
Pr(B∩I)=0.40(0.20)=0.08
Pr(F∩I)=0.60(0.50)=0.30
Compute
Compute
Normalize the broad-mode contribution by the total probability of the reported interval.
Pr(B∣I)=0.08+0.300.08=194=0.2105263…
Answer
Answer
Given the interval reading, the probability of broad mode is approximately 0.211.
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.