Financial Simulation Modeling in Excel

UNDERSTANDING SIMULATION RESULTS

Now that we have the tools to run simulations and had practice in specific situations, we need some rudimentary understanding of how to interpret our results. As we talked about earlier in Chapter 1, the need to perform a simulation could arise from the fact that the result cannot be explicitly determined. However once the simulation is performed, we need to ask ourselves, “How much can we trust the outcome?” It would be rather irresponsible for the practitioner to run a simulation without any thought given to the accuracy or trustworthiness of the result and then present the outcome as a statement of fact. First we must understand what an error is and the various types that we can encounter throughout our analysis.

Statistical Error

Earlier in this chapter we introduced the N-Sided Die example. While the result of the exercise might seem trivial, it presents many useful concepts that are applicable to all simulations in general when discussing errors. When dealing with random processes such as rolling a die, the main source of error is statistical. Imagine for a moment we flip a coin once and it lands on heads. That one sample tells us that the probability of landing on heads is 100 percent; however, this is clearly wrong since, as we all know, the probability is 50 percent. After five such flips we may get a set of results that may look like the following {H, H, T, H, T}, which implies that the probability of landing heads is now 60 percent. As ...