Friday, November 10, 2017

Monte Carlo Simulation: Assessing A Reasonable Degree of Certainty

Monte Carlo Simulation: Assessing  A Reasonable Degree of Certainty
A summary By Kevin Muvunyi


In their article “Monte Carlo Simulation: Assessing A Reasonable Degree of Certainty”, Daily and Solis apply the Monte Carlo simulation technique to two hypothetical scenarios that seek to determine future financial outcomes with a certain degree of confidence by examining the benefits and drawbacks of the methodology. According to the authors, Monte Carlo simulation has a large scope of applicability in various fields and more so in financial analysis hence the interest therein. 

In their analysis, Daily and Solis examine a simple lost profits analysis and then a more complex construction delay claim requiring the evaluation of lost profits. First, they begin by tackling each scenario based on known facts, evidence, and assumptions followed by a repeat of the same process but this time using the Monte Carlo simulation. In the case of the lost profits analysis, the researchers first utilize single inputs as part of their assumption based analysis to get the lost profits values. They then proceed to use the Monte Carlo technique with the help of the Microsoft Excel based RISK program, whereby with the use of probability distributions they are able to run 10000 iterations to get final results. What the authors were able to discern in this particular case is that the lost profits value in both instances were approximately similar. In the second case of the construction delay claim, the researchers repeated the same processes but this time around due to the complexity of the scenarios they were inclined to use multiple inputs, thus, there was significant material differences between the results of the two methodologies, namely the Monte Carlo simulation and the assumption based technique. Ultimately, Daily and Solis conclude that the Monte Carlo simulation is able to have a material effect on the ultimate outcome or no material effect at all in regards to financial analysis. Nonetheless, they stress the fact that in both scenarios Monte Carlo provided them with helpful statistics regarding the possible outcomes of their analyses.


Although the article provides two practical examples of how the Monte Carlo simulation can be applied to real world scenarios it nonetheless fails to clearly demonstrate the drawbacks of the technique in a financial analysis context. 



  1. How much did the additional variables affect the accuracy of the Monte Carlo simulations? Similarly, how did they adjust for the assumption model not really have standardized values for the more complex scenario?

  2. Also, what numbers did they use for the single digit inputs? If they're using random numbers for the variables which variables are they representing?

  3. I think it would have been more interesting to run this study comparatively to a different technique in order to show the benefit of a monte carlo simulation.

  4. I have something special for you

    This Profiteering Software App usually
    retails at $497 but for my subscribers
    it’s FREE!

    This is invite only: