Chapter 9: Risk Analysis Flashcards
When will a decision involve risk?
A decision will involve risk where the possible values could have a significant impact on a project’s profitability
The estimation of the probability of an event usually involves…
Subjectivity
In risk analysis, different forms of subjectivity need to be addressed, in deciding:
- What the degree of uncertainty is;
- Whether the uncertainty constitutes a “significant risk”;
- Whether the risk is acceptable
Sensitivity Analysis
Estimating the extent to which the outcome is sensitive to the assumed values of the inputs
Two steps in performing a sensitivity analysis
- Establish which of the various input variables impact most on the outcome (NPV or BCR)
- Undertake and report the range of results allowing each of these to vary between low and high (pessimistic and optimistic) - individually and in combination
Rule of thumb for variation
There is no golden rule about how much variation around the “best guess” estimation should be allowed, but 20% is a good rule of thumb
Important information learned from a sensitivity analysis
Whether or not the NPV or BCR of the policy option could be negative under some scenarios within a reasonable range of assumed input values
However, does not tell us the likelihood of this happening
Applied risk analysis
Use of discrete probability distributions to compute expected value or variable rather than a point estimate
Applied risk analysis: joint probability distributions
We are usually uncertain about more than one input/output
The probability distribution for NPV depends on aggregation of probability distributions for individual variables
Joint probability distributions can be for correlated and uncorrelated variables
Continuous probability distribution: e.g. the normal distribution
Represented as a bell-shaped curve
This distribution is completely described by two parameters:
- the mean
- the standard deviation
Degree of dispersion of the possible values around the mean is measured by the variance (s^2) or the standard deviation (s)
Expected Wealth
(W1 x p1) + (W2 x p2)
Expected Utility
p1 x U(W1) + p2 x U(W2)
Risk modelling in Excel: Monte Carlo simulations
In Monte Carlo simulations we use additional software to perform a formal risk analysis
In the simulation the project’s net benefits are recalculated thousands of time using random values for input variables from a given distribution
It assembles the results and presents them in the form of a probability distribution, showing the likelihood of achieving a given outcome
Suppose NPV>0. Is there any reason (other than budget constraint) why you would recommend that the project should not go ahead immediately?
Uncertainty and the value of information
Might be uncertainty about the values of some of the variables used to calculate the NPV (e.g. future prices)
Delaying the project might resolve these uncertainties
How to investigate the value of delaying the project
Compare the NPV of undertaking the project immediately at time 0 with the NPV of delaying the project until time 1