As the development of a trading strategy progresses, the analysis also progresses, and the performance hurdles a strategy must meet to be considered viable get more stringent. My primary method of analysis at the later stages is Monte Carlo analysis. But before I explain how I run the analysis and what I look for in results, I’ll first briefly describe the process.
Monte Carlo analysis, or simulation, sounds like a daunting topic, but actually it is not. With the Monte Carlo spreadsheet I created, which you can download for free (www.wiley.com/go/algotradingsystems), the analysis is pretty simple. But what is it actually?
Think about the individual trades in your strategy. These trades taken sequentially, in the order they occurred, yield the strategy equity curve. But what if the order of those was different? Could the drawdown become more severe? Could the end equity be different? These are the questions Monte Carlo analysis can answer.
In its simplest form, you can think of it this way: First, get a number of little pieces of paper, one for each trade in your strategy. Then, write down one trade result on each piece of paper. Once you have all trades accounted for, put all the pieces in a hat. Randomly choose one. That is your first trade. Record it, adding it to your initial equity, and then put the piece of paper ...