13.1 Risk and uncertainty in the short term Flashcards
Risk and uncertainty both refer to the likelihood
- That the outcome from decisions may differ from those which are expected when a decision is taken
- Decision makers will aim to account for both in decision making
We tend to differentiate between risk and uncertainty in terms of availability of probabilities attached to each outcome
- Risk is quantifiable - Potential outcomes of decision are predictable and the probabilities for each possible outcome can be estimated
- Uncertainty is unquantifiable - Decision maker may not be able to estimate the likely outcomes and/or cannot estimate the probability that outcome will occur
Dealing with ……. is more challenging / Potential approaches
- Uncertainty is more challenging than dealing with risk
- Potential approaches are to reduce time horizon for decisions (tends to be more uncertainty over longer time frames) or attempting to convert uncertainty into risk (eg market research)
- There are mathematical models available for dealing with risk
Expected value
- Is one way to account for risk by determining an expected value based on the range of possible outcomes
- An expected value condenses all the different possible outcomes into one overall average result by calculating a single weighted average
- It is the most likely result, even if not a possible result (finds average if same event took place a 1000 times)
Expected value formula
EV = Sum of (px)
x = Future outcome
p = Probability of outcome occurring
Limitation of expected value technique
- It assumes that the decision maker is risk neutral
- A risk neutral investor neither seeks risk or avoids it, they are happy to accept an average outcome
- But decision makers will not normally be neutral and therefore pay no attention to the expected value
Three main types of decision makers
- Risk neutral - Consider all possible outcomes and will select strategy that maximizes the expected value or benefit, therefore focus attention on expected value
- Risk seekers - Likely to select strategy with best possible outcome regardless of likelihood they will occur, therefore ignore the expected value
- Risk averse - Try to avoid risk. They would rather select a lower but certain outcome than risk going for a higher payoff which is less likely to occur. Therefore also ignore expected value
Advantages of expected values:
- Takes account of risk
- Easy decision rule
- Simple
Disadvantages of expected values:
- Subjective
- Not useful for one offs
- Ignores attitudes to risk
- Answer may not be possible
- Ignores the spread of outcomes
Pay off tables
- A table that illustrates all possible profits / losses
- To consider the risk borne by each alternative it is necessary to consider all these different profit / loss possibilities
- Managements decision will often depend upon their attitude to risk
The standard deviation compares
- All the actual outcomes with the expected value (or mean outcome) and then calculates how far on average the outcome deviates from the mean using a formula
- Helpful to show how wide ranging the possible outcomes are
- It is a measure of volatility and the more variation the more volatile the returns and therefore the more risk involved with decision
Standard deviation formula:
SD = Square root of [Sum of (X - Xmean)sq] / n
X = Each value in data set
n = Number of values in data set
Coefficient of variation measures
- The relative size of the risk for projects that have very different std deviations
- The smaller the coefficient of variation the less dispersed the variable is and therefore the less risky
CoV = SD / mean
Using normal distributions in decision making
- If we know the mean and the std deviation for a distribution we can work out the probability (% chance) of a certain value occurring
- We first convert normal distribution to a std normal distribution
- Which as a mean of 0 and a SD of 1
Std normal distribution formula
z = x - u / SD
x = variable
u = mean
SD = Std deviation
The z score allows us to calculate the proportion of distribution meeting certain criteria for any normal distribution (ie determine possibilities)