5 Monte Carlo simulations

 

This chapter covers

  • Why simulations are important for financial planning
  • The difference between arithmetic and geometric average returns
  • Estimating the probability of running out of money in retirement
  • Implementing the bootstrapping method
  • Incorporating inflation and longevity risk

Monte Carlo simulations have numerous applications in wealth management and financial planning. In this chapter, we will focus on a specific problem that is particularly well-suited for Monte Carlo analysis: whether you will run out of money in retirement.

In Monte Carlo simulations, random scenarios are generated and analyzed. Most people focus on the randomness of stock and bond returns, but Monte Carlo simulations can incorporate anything random, like inflation, health care expenses, life expectancy, or even future tax rates.

Why are Monte Carlo simulations necessary, rather than simply making projections based on expected outcomes? For example, if you have a 50-50 mix of stocks and bonds and expect your portfolio to have average returns of 5%/year (say, 8% for stocks and 2% for bonds), couldn’t you just project how your savings are expected to evolve over time to figure out whether your money will last through retirement? This is referred to as a straight-line projection. There are several problems with that approach:

5.1 Simulating returns in Python

5.2 Arithmetic vs. geometric average returns

5.3 Simple vs. continuously compounded returns

5.4 Geometric Brownian motion

5.5 Estimating the probability of success

5.6 Dynamic strategies

5.7 Inflation risk

5.8 Fat tails

5.9 Historical simulations and booststrapping

5.10 Longevity risk

5.11 Flexibility of Monte Carlo simulations

Appendix

Summary