Risk and decision making Flashcards
(19 cards)
Methods of dealing with decision making under risk and uncertainty
Setting a minimum payback period
Increasing the discount rate to use a higher hurdle rate and get a more conservative NPV
Calculating a range of outcomes
Sensitivity analysis
Sensitivity analysis
Determines how sensitive the NPV of a project is to an individual estimated variable
Shows the % change in the variable necessary to change our decision
Sensitivity to variables impacting on cashflows
NPV of project / PV of cashflows impacted by the variable
Sensitivity to the cost of capital
(IRR - Cost of capital) / Cost of capital
Sensitivity analysis pros and cons
Facilitates decision making
Identifies critical variables
Simple to understand
vs
Assumes variables change independently
Ignores probability
No clear answer - only gives context to NPV calculation
Predictive analytics
Monte Carlo simulation
Linear regression models
Prescriptive analysis
Monte Carlo simulation
Identifies each of the different variables, the range of the different values
of those variables and the probabilities of those values occurring
Runs simulations to record the NPVs of the project for different combinations of values for the different variables
Results show the expected NPV and the distribution of possible NPV values.
Monte Carlo simulation pros and cons
Gives more info about spread of possible outcomes
Useful for problems which cannot be solved analytically
vs
No clear answer - only gives context
Time consuming without computer
Expensive to design and run
Assumptions made about the probabilities associated with different variables
Linear regression models
Quantifies the relationship between the dependent variable and the independent
variables
Linear regression pros and cons
Simple to use and easy to explain
Predict the impact of expanding variables beyond current estimates
vs
Not always be a linear relationship between variables and outcome
More complex models needed to consider additional variables concurrently
May identify false relationships between variables and outcomes - does not consider correlation vs causation
Correlation coefficient formula
=CORREL
Prescriptive analysis
Calculates the optimum outcome from a variety of business decisions such as capital rationing, replacement analysis or balance of finance
Prescriptive analysis pros and cons
Identify optimum investment decisions whilst considering the impact of multiple decisions and variables
vs
Creating reliable prescriptive models is complex
Reliability depends on reliability of data used
Types of data bias
Selection Bias
Self-selection
Observer
Omitted variable
Cognitive
Confirmation
Survivorship
Statistical tools
Mean (=AVERAGE)
Standard deviation (=STDEV)
Coefficient variation - measures the standard deviation as a % of the mean
Normal distribution
Pros and Cons of expected values
Info reduced to a single number for each choice
Easily understood
vs
Probabilities may be difficult to estimate
May not equate to a possible outcome
Represents long-run average - inappropriate in one-off situations
Gives no indication of the spread of possible results
Portfolio theory
It is possible to diversify away exposure to specific risk by holding shares in a diverse selection of different companies
Holding a large portfolio eliminates the majority of exposure to specific risk, leaving only exposure to systematic risk
Capital Asset Pricing Model
Measures systemic risk of investments by determining the required rate of return (cost of capital) on an investment
CAPM pros and cons
Directly links the risk associated with an investment to the required return from that investment
vs
Assumes investor diversified
Assumes investors can deposit and borrow at risk-free rate
Assumes exposure to all systematic risk can be captured in one measure (β) that is accurately measured and unchanged
Uses historic figures in the calculation
Ignores potential agency problem if directors are not diversified
