Monte Carlo Simulation as a Risk Management Tool Key Terms
Don't play games with your business; use Monte Carlo simulations to manage risk
Atom-bomb scientists coined the term Monte Carlo method after Europe’s casino capital in Monaco to highlight the "random chance" aspect of the model. Monte Carlo simulations perform multiple calculations using random variables to determine the probably outcome of a given situation. Today specialists employ the technique in fields as diverse as weather prediction, oil exploration, sales forecasts and financial planning. Before using Monte Carol simulation models or working with a consultant to manage risk, get familiar with some key terms to understand Monte Carlo basics.
Stochastic
Mathematicians refer to Monte Carlo simulations as stochastic, meaning they involve conjecture or randomness. This refers to the way a simulation model generates random numbers to pose possible outcomes, as well as calculate the likelihood of these outcomes occurring. At the opposite end, with "deterministic" you're in control by calculating one set of "if/then" answers for each "what if" question you pose on a set of parameters.
Try: TreeAge Software discusses the use of parameters both in pre-set, deliberate (or deterministic) analysis as well as random (or stochastic) analysis.
Parameters
Parameters are the boundaries you place on the model--what factors will affect the outcome that you're trying to test, and for each factor, what is its range is from absolute minimum to absolute maximum. When you set these parameters, the Monte Carlo system selects random numbers within those defined ranges to run through the algorithm to calculate the final solution set.
Try: Beige Bag Software demonstrates how you use parameters to set up or change the Monte Carlo simulation model.
Algorithm
If you've seen a flowchart, you've looked at an algorithm that's been graphed out. It's the frequently complex mathematical instructions a Monte Carol simulation package uses to generate the random numbers and perform the calculations necessary to create the model.
Try: The bottom of the Michigan State University web page delineates the components comprising a typical Monte Carlo algorithm.
Sensitivity
All things aren't equal in the real world; some aspects of a business scenario carry more weight on the circumstances you're evaluating. The outcome you want to model can be extremely sensitive to certain aspects, immune to others. The Monte Carlo simulation correlates the individual parameters' measured impact to the forecasted outcome, providing you with a chart showing how much impact any given parameter has on the overall outcome.
Try: Investopedia explains sensitivity analysis in financial analysis, such as stocks and bonds. Pharmalicensing shows the output of a Monte Carlo simulation, including a sensitivity chart.
Worst case
Monte Carlo simulations, based on calculations from random numbers, helps you build a model of uncertainty to better understand business situations. Rather than hoping for the best and fearing the worst outcomes, ask "what if" with the Monte Carlo simulation to determine best-case and worst-case outcomes, plus what probability the best-case or worst-case scenario will occur.
Try: Frontline Systems' Solver.com posts a step-through of Monte Carlo simulations, describing how they model worst-case scenarios. The Materials Research Society posts an abstract discussing the use of Monte Carlo modeling on the Yucca Mountain nuclear waste repository to determine worst-case scenarios for the barriers’ lifespan.
Stress test
If you want to know how your business will fare under adverse conditions, use a Monte Carlo simulation to create a stress test. The test will create a baseline for you to better assess overall risk by answering "how bad can it get?"
Try: Palisade Corporation talks about how the Federal Reserve tested the nation’s banking system after the housing market meltdown.
Copyright © 2011 Business.com, Inc. All Rights Reserved.
