ACAP Cost Risk and Uncertainty Analysis 3

Question 1

What is 'program uncertainty' as distinguished from 'estimating uncertainty' in cost risk analysis?

Accepted Answer

Program uncertainty stems from unknown future events; estimating uncertainty stems from imprecision in the cost model itself

Answer

Program uncertainty is always larger than estimating uncertainty

Answer

Program uncertainty applies only to EMD phase programs

Answer

Estimating uncertainty is irreducible while program uncertainty can be managed

Question 2

Which term describes the practice of assigning probability distributions to cost estimating relationships (CERs) to reflect statistical uncertainty in the regression?

Accepted Answer

Parametric risk analysis

Answer

Analogy risk analysis

Answer

Regression bootstrapping

Answer

Cost element uncertainty mapping

Question 3

In DoD cost risk analysis, what is the 'most likely cost' (MLC)?

Accepted Answer

The mode of the cost probability distribution

Answer

The arithmetic mean of all cost scenarios

Answer

The budget authority approved by Congress

Answer

The 50th percentile of the S-curve

Question 4

What is a 'risk register' and how does it relate to cost risk analysis?

Accepted Answer

A documented list of identified risks with their likelihood, impact, and mitigation strategies used to inform cost uncertainty ranges

Answer

A list of all past cost overruns on similar programs

Answer

A database of contractor-reported risks during source selection

Answer

A schedule of risk review board meetings

Question 5

Why is a log-normal distribution often preferred over a normal distribution for cost element uncertainty?

Accepted Answer

Because costs cannot be negative and log-normal is bounded at zero with a right-skewed tail

Answer

Because log-normal distributions have narrower tails, reducing overrun probability

Answer

Because DoD mandates log-normal distributions in all cost risk models

Answer

Because log-normal distributions are easier to correlate in Monte Carlo simulations

Question 6

What does 'normalizing' data mean in the context of preparing inputs for a cost risk model?

Accepted Answer

Adjusting historical cost data to remove the effects of inflation, quantity, and other non-recurring factors before using it to set uncertainty ranges

Answer

Converting costs to a standard normal distribution

Answer

Scaling all cost elements to the same order of magnitude

Answer

Removing outlier data points from the dataset