Risk & Uncertainty Flashcards
What is the difference between risk and uncertainty?
Risk is quantifiable. There will be a number of possible outcomes and the probability of ech outcome is known based on past experience. This allows the decision maker to take risk into account through the use of mathematical techniques.
Uncertainty is unquantifiable. There are a number of possible outcomes but the probability is not known. As a probability cannot be assigned to each outcome they cannot be mathematically modelled and it is much harder to account for uncertainty when making a decision.
What are expected values?
Expected values are a mathematcial technique that allow you to quantify the effect of risk on all possible outcomes from a decision. It is calculated as the sum of all possible outcomes, weighted by the probability of that outcome happening. It is the long run average (mean), It is not the most likley result or even a possible result, instead it is the average result if the decision was made multiple times.
What are the benefits of expected values?
- When the outcome of a project isnt known it is useful to take into account many possible outcomes
- It takes into account risk because it uses probabilities to weight each outcome
- The decision information is reduced to and represented by a single number which helps to facilitate decision making
- The calculations are relatively straightforward
What are the drawbacks of expected values?
- The probabilities are usually subjective - two individuals will not necessarily assign the same probability to a particular outcome.
- The expected value does not necessarily correspond to any actual possible outcome. They are therefore not useful or reliable for one-off events.
- It ignores risk attitudes and assumes the decision maker is risk neutral.
- It does not take into account non-financial factors such as quality or reputation
- They do not take into account the range of values that could occur. There may be two scenarios with the same expected value, but one may have a much wider dispersal of outcomes and therefore a higher risk.
What is standard deviation?
Standard deviation measures how much the possible outcomes vary compared to the average, which is the expected value.
This is useful because expected values dont take into account the range of values or the difference between the highest and lowest levels and the risk associated with this.
The higher the standard deviation the higher the risk.
How do expected values relate to risk attitude?
Expected values asume a neutral risk attitude.
What is the coeffient of variation?
This is the standard deviation divided by the expected value. It takes into account both the expected value and risk as measured by the standard deviation. So, in terms of decision making, the optimal decision would be to choose the option with the lowest coefficient of variation.
What is the maximax rule?
The maximax rule involves selecting the decision which would maximise the maximum pay-off potentially available. It is based on the assumption that the best pay-off will occur and optimistically ignores any probability of it not occuring.
This is the option suitable for a risk seeking decsion maker.
What is the maximin rule?
The maximin rule selects the best possible option from the worst possible outcome. It is a more pessimistic view assuming the worst will happen and then picking the best option if it does.
This is the option suiatble for risk adverse decsion makers.
What is the minimax regret strategy?
The minimax regret strategy looks to minimise the maximumm regret. Regret in this context is defined as the opportunity loss through having made the wrong decision. It is found by calculating the difference between the outcome of the best option in the scenarion and the outcome of that particular decision.
This is also a risk adverse solution, suited to ‘sore loosers’ or decision makers looking to avoid a wrong decision.
What is a pay-off table?
A pay-off table is often used to illustrate all of the possible profits and losses that could occur because of a decsion. It can be a useful way to represent and analyse a scenario in condition of uncertainty where there is a range of possible outcomes and responses.
The option chosen within the pay-off table would depend on the decision makers risk appetite.
What is a Decision Tree?
Decision Trees are diagramatic representations of a decision problem. It is used in situations where mutiple decisions are to be made that follow on from one another.
The branches represent all the possible courses of action and the possible outcomes from these actions.
How do you read a decision tree?
- A square represents a decision point.
- A circle represents an outcome point.
- The branches from the outcome points are always subject to probabilities
- You should evaluate the tree from right to left, calculating the EV at each outcome point and then choosing the best decsion (with highest EV) at each decision point.
What are the benefits of decision trees?
- Clearly maps out all the decisions and uncertain events and shows how they are interrelated
- Are particuarly beneficial where the outcome of one decision affect another decision
What are the disadvantages of decision trees?
- It is based on expected values and thus suffers the limitations of this technique.
- They are often oversimplified representations of the sitaution in order to make them manageable. This makes the decision appear far more discrete than it actually is.
