This Exam P sample reference tests Normal Distribution. This is a sum of independent normal variables. Their aggregate has mean 300 and standard deviation √(290000), so its positive-tail probability is 0.71127 and choice E is correct.
These notes identify the calculation error associated with each wrong letter when that error is reproducible.
AA value below one half reverses the sign or tail despite the aggregate mean being positive. The probability of a positive result must exceed one half here.
BThis understates the signal-to-noise ratio, often by retaining only the magnitude 100 when combining the two means. Signed means must be added before standardization.
CThis is consistent with using an overly large scale after blending means and standard deviations. Only variances, not unlike quantities, can be combined in the variance calculation.
DAdding the standard deviations directly gives 700 and Φ(300/700)=0.6666, which rounds to 0.67. Independent standard deviations combine through their squared values.
Original practice · fully worked
Original variant: paired sensor correction
A materials lab applies two independent normally distributed sensor corrections. Correction A has mean 12 units and standard deviation 3 units; correction B has mean -5 units and standard deviation 4 units. The final adjustment is D=A+2B. Find the probability that D is negative.
A 0.081
B 0.407
C 0.428
D 0.500
E 0.593
Variant answer in brief
The linear combination has mean 2 and variance 73. Standardizing zero gives Φ(-2/√(73))=0.40746, so choice B is correct.
Setup
Setup
Represent the final adjustment as the stated linear combination of the two independent inputs.
D=A+2B
Model
Model
A linear combination of independent normals is normal. Apply the coefficient to both the mean and the variance contribution.
E[D]=12+2(−5)=2
Var(D)=32+22(42)=73
Compute
Compute
Standardize the zero threshold under the resulting normal distribution.
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.