I want to just briefly go through Monte Carlo simulation just to talk
about this particular tool that is widely used as a way to
model especially uncertainties and to help make decisions.
So the way we do this,
we use probability distributions to represent uncertainty on model,
parameters or variables or inputs.
Now what we're going to do is,
we're going to run through many many times on each trial or each iteration,
we're going to reach into each of those probability distributions and
pull out a random number according to that probability distribution.
Then we'll use our model to calculate the values of the outputs that we hear about,
those consequences we care about,
and we'll get a so-called risk profile that represents
the probability associated with different possible outcomes.
When you do this,
you get this nice picture of the risk that you face in what might have been
a very complicated problem where
the complications came from many different sources of uncertainty.
So what can Monte Carlo simulation tells us?
Well, it can help us identify key uncertainties.
By understanding those output risk profiles,
we can get an idea of things like the probability of losing money,
of achieving a goal,
we could get a picture of the expected outcomes.
It's very clear in making a decision in
a case like this when we're using Monte Carlo simulation.
That making a decision means choosing a gamble that is
appropriate for the decision makers particular risk appetite.
We can also look for better options.
We can look for ways to reduce risk, and of course,
we can do that with a decision tree too and we should.