Welcome to the third of five (sort of) posts about Monte Carlo risk analysis using Excel VBA. Actually, what was going to have been today’s post has turned out to be much longer than my target of around 3,000 words, so I have decided to split it in two, with the second half appearing tomorrow.
Yesterday we looked at how to generate arrays of random numbers using a combination of inverse transform and ‘latin hypercube’ sampling. Today we get to the heart of the problem and consider how to induce correlation onto our random sample using the Iman-Conover method. This is based on the idea of rank-order correlation. It involves the following steps: Read more