Know these first
- L3.2.11: Monte Carlo VaR
Math Atoms
By the end you will be able to
- 1Price the desk's puts by simulating the index under the risk-neutral measure, and give the estimate with its standard error and a 95% interval.
- 2Use the 1/√N law to plan the number of draws, and compare plain, antithetic and control-variate estimates.
- 3Price an arithmetic Asian put by simulation, with the geometric Asian as a control variate.
- 4Price a daily-monitored down-and-out put by simulation, and correct the continuous formula for discrete monitoring.
Why it matters
Black-Scholes prices the desk's puts in a microsecond, so why simulate? Because the formula is the exception. Change the put so that it pays on the average of the index over three months, or so that it dies if the index ever closes below 7,000, and there is no formula, or only one for an idealised version. Simulation prices almost anything whose payoff can be written down. The price of that generality is noise: every Monte Carlo price comes with an error bar, and this lesson is about making it small.