Lecture 10 Flashcards
purpose of risk measurement/assessment
- understand the dimensions of ERM problems quantitatively
- measure the risk-return trade-off of ERM alternatives (tools)
- predict/forecast outcomes to the best of our ability (reduce risk)
random variable
don’t know which value will occur (can have multiple)
collectingt data about random variables
- what do we want to know?
2. where can we find info/data to learn what we want to know?
develop probability distributions
- a method of summarizing a large collection of data
- identify all possible outcomes
- calculate probabilities for all possible outcomes
identify all possible outcomes
- distinguish possible from observed/actual outcomes
- collectively exhaustive
- mutually exclusive
collectively exhaustive
define so that the occurrence of one outcome precludes occurrence of any other
mutually exclusive
account for all possibilities
general characteristics of probability distributions - quantitatively
- central tendencies
- dispersion
central tendency
- expected value
- median
- mean
dispersion
- range
- variance
- standard deviation
- coefficient of variation
- value at risk
- maximum probable and maximum possible loss
uncorrelated outcomes/events
- aka independent
- the occurrence of one event does not tell us anything about the probability of the other
- the occurrence of “a” does not tell us anything about the probability of “b”
correlated outcomes
- aka dependent
- the occurrence of one tells us something about the probability of the other
- the occurrence of “a” does tell us something about the probability of “b”
where do we find data?
- past docs
- own/industry loss history
- theory
- how valuable is this extra data? (because it can be costly)
when to use coefficient of variation
when you don’t have the same mean
critical probability
probability so low we don’t really care about it (cut out of distribution)
