Monte Carlo analysis is like a map that hasn’t been fully filled in. You can see where you have been, you can see where you want to go, but the details on how to get there aren’t fully fleshed out.
From a mechanical perspective, Monte Carlo analysis is a statistical technique to estimate probabilities of future outcomes in an uncertain system. The analysis uses a large number of simulated paths, or runs, to estimate probable outcomes. In our case we run 1,000 simulations of your clients’ financial plan, and then aggregate the results to generate a probability of success.
To generate a simulated path there are three basic inputs for each year:
- The portfolio value
- The net cash flow for the year
- The annual rate of return
We know the portfolio value because we either just generated the value from a previous year, or it comes from the initial values in your client’s financial plan. We know the net cash flow from the contributions and goals that you have entered for the plan. The unknown input is the annual rate of return.
To determine the annual rate of return we create a normal curve from the average return and standard deviation of the model portfolio selected for that year. We then randomly select a return from that curve.
Once we have all three inputs for the year we compute the portfolio value at the end of the year and move onto the next year. Once we reach the end of the plan we move onto the next simulated path.
To generate the plan’s probability of success, we simply look at how many of the 1,000 simulations ran out of money. If a simulated path meets all of the goals it is a success. If it does not meet all of the goals in the plan it is not successful. The plans probability of success is the percentage of the simulated paths that were successful.
Monte Carlo analysis is a club, not a scalpel. Use it as a sanity check for your client’s financial plans, but it is not a panacea. There are limitations to Monte Carlo analysis; nonetheless, it is still an extremely valuable tool.
Comments