This Exam P sample reference tests Binomial Distribution. The two treatment counts are independent binomials with variances 1.35 and 3.6. Squaring the benefit coefficients gives total variance 4(1.35)+9(3.6)=37.8, choice B.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AThe value 13.5 is 2(1.35)+3(3.6), obtained by scaling variances with payment coefficients rather than their squares.
CThe value 108 is 15[4(0.9)+9(0.4)]. It treats weighted raw Bernoulli second moments as variances and omits the (1-p) factors.
DThe value 202.5 is 15²(0.9), a raw second moment from treating the whole radiation group as one all-or-nothing block; it is not the aggregate variance.
EThe value 567 is 15(37.8), multiplying by the group size a second time even though both binomial variances already include 15 trials.
Original practice · fully worked
Original variant: aggregate indicator variance with dependence
Three event indicators I1, I2, and I3 each have event probability 0.20. Every pair has covariance 0.02. A payment of 100 is made for each occurring event. Calculate the variance of the aggregate payment.
A 0.60
B 1600
C 4800
D 5400
E 6000
Variant answer in brief
The indicator-count variance is 3(0.16)+2(3)(0.02)=0.60. Multiplying by 100² gives aggregate payment variance 6000, choice E.
Setup
Setup
Write the aggregate as 100 times the sum of three indicators.
T=100(I1+I2+I3)
Var(Ii)=0.2(0.8)=0.16
Model
Model
Include each of the three pairwise covariances twice in the variance of a sum.
Var(I1+I2+I3)=3(0.16)+2(23)(0.02)
Compute
Compute
Evaluate the count variance and apply the squared payment scale.
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.