This Exam P sample reference tests Exponential Distribution. Conditioning on exceeding 100 shifts the exponential excess distribution. Its conditional 95th percentile is 100-300 log(0.05), about 999, so choice E.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
DThe value 900 is the rounded unconditional 95th percentile, about 898.7. It omits the 100 units already survived.
Original practice · fully worked
Original variant: conditional upper quantile of a queue delay
Queue delay T is exponential with mean 40 minutes. Given that a customer has already waited more than 15 minutes, find the 80th percentile of the customer’s total wait.
A 64.38 min
B 72.44 min
D 79.38 min
C 86.44 min
E 95.00 min
Variant answer in brief
The remaining wait is exponential with mean 40. Adding its 80th percentile, -40 log(0.2), to 15 gives 79.38 minutes.
Setup
Setup
Given a wait beyond 15 minutes, memorylessness makes the additional wait exponential with mean 40.
T−15∣T>15∼Exp(mean 40)
Model
Model
The 80th percentile of total wait is 15 plus the 80th percentile of the additional exponential wait.
P(T≤q∣T>15)=1−e−(q−15)/40=0.80
Compute
Compute
The additional-wait percentile is 64.3775 minutes; adding the elapsed 15 gives 79.3775 minutes.
q=15−40log(0.20)=79.3775
Answer
Answer
Therefore the conditional 80th percentile of total wait is 79.38 minutes, corresponding to choice D.
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.