This Exam P sample reference tests Normal Distribution. The relevant linear combination is normal with mean 1 and variance 50.49. Standardizing zero gives a probability of 0.55596, which selects choice C.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 0.407 is the complement of 0.593. It results from both treating the third lifetime as fixed and reversing the requested comparison.
BThe value 0.444 is the complementary normal probability P(W<0), approximately 0.44404.
DThe value 0.593 is obtained by treating 1.9 times the third lifetime as the fixed value 19. That leaves variance 18 for the first two lifetimes and gives Φ(1/√(18))=0.593.
EThe value 0.604 corresponds to a still smaller standard error. All three independent lifetime variances contribute to W.
Original practice · fully worked
Original variant: percentile of a calibrated Gaussian score
A laboratory models two independent calibration offsets: U has a Gaussian distribution with mean 1 and variance 4, while V has a Gaussian distribution with mean -2 and variance 9. The reported composite score is T=3U-2V. Determine its 95th percentile.
A 7.000
B 16.009
C 20.957
D 23.631
E 26.738
Variant answer in brief
The composite score has mean 7 and variance 72. Adding 1.64485 standard deviations to its mean gives the 95th percentile 20.957, choice C.
Setup
Setup
Compute the center of the reported linear score.
E[T]=3(1)−2(−2)=7
Model
Model
Square the two coefficients when propagating independent variances.
Var(T)=32(22)+(−2)2(32)=72
T∼N(7,72)
Compute
Compute
Use the one-sided standard-normal 95th percentile.
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.