This Exam P sample reference tests Exponential Distribution. This is a comparison of two independent exponential amounts. A one-sided ratio event has probability one-third, and the two symmetric directions are disjoint, so their total probability is two-thirds and choice E.
How to solve this Exponential Distribution question
Setup
Setup
Let X and Y denote the two amounts. Their common mean is one, so each has rate one.
fX(x)=e−x,Pr(Y>t)=e−t,x,t≥0
Model
Model
First evaluate the direction in which Y exceeds twice X. The opposite direction has the same probability, and the two strict inequalities cannot occur together.
Pr(Y>2X)=∫0∞Pr(Y>2x)fX(x)dx
Compute
Compute
Insert the exponential survival function and density, then use symmetry.
Pr(Y>2X)=∫0∞e−2xe−xdx=∫0∞e−3xdx=31
Pr(Y>2X or X>2Y)=2(31)=32
Answer
Answer
The required probability is two-thirds.
32(E)
Continue without hunting through PDFs
All 718 Exam P sample solutions in syllabus order
The searchable 3108-page manual includes this complete solution, its error analysis, and one original worked variant for every active reference.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 1/6 results from taking the correct one-direction probability 1/3 and multiplying by 1/2 for symmetry. Symmetry creates a second disjoint direction, so it calls for addition rather than another restriction.
BThe value 1/4 is the incorrect integral of exp(-4x), obtained by using exp(-2x) for both the conditional tail and the conditioning density. The actual conditioning density is exp(-x), so the combined exponent is three.
CThe value 1/3 is the probability of only one direction, such as Y>2X. The equally likely disjoint event X>2Y must also be included.
DThe value 1/2 is the probability that a specified amount is merely the larger one. It ignores the stronger requirement that the larger amount exceed twice the other.
Original practice · fully worked
Original variant: latency inside a directional band
A data job has independent setup latency S with exponential rate 2 and transfer latency T with exponential rate 3. A monitoring flag appears when the transfer latency is greater than the setup latency but less than three times the setup latency. Calculate the probability of a flag.
A 0.1818
B 0.2182
C 0.4000
D 0.5818
E 0.6000
Variant answer in brief
The flag is the part of the tail T>S that remains after removing T>3S. Exponential tail averaging gives 2/5-2/11=12/55=0.2182, so choice B.
Setup
Setup
Write the two independent exponential models with their unequal rates.
fS(s)=2e−2s,Pr(T>t)=e−3t,s,t≥0
Model
Model
The upper-tail event T>3S is nested inside T>S, so subtract it from the broader tail.
Pr(S<T<3S)=Pr(T>S)−Pr(T>3S)
Compute
Compute
Average each transfer tail over the setup density.
Pr(T>cS)=∫0∞e−3cs2e−2sds=2+3c2
Pr(S<T<3S)=52−112=5512=0.2181818…
Answer
Answer
The monitoring flag appears with probability about 0.2182.
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.