Upcoming Courses

June 7-9, 2006
(early registration with 10% discount ends May 12th!)
Introduction to quantitative risk analysis
Princeton, NJ

July 12-14, 2006
Corporate risk analysis (in Spanish)
Medellнn, Colombia

Risk Analysis Tip (a sample ModelAssist topic)

Presenting your Risk Analysis Model and Results (2 of 3)

The following Risk Analysis tip has been drawn from material in ModelAssist®, Vose Consulting's risk analysis training, reference and template software. ModelAssist users can consult the ModelAssist-references (in the form of Mxxx) for additional information. To read more about ModelAssist and download the free demo version, go to http://www.voseconsulting.com/software.htm.

Introduction

In the last Risk Analysis Tip, we explained the importance of guiding the reader or you risk analysis report or presentation guides the reader through the assumptions, results and conclusions in a manner that is transparent, efficient and interesting. While this last Risk Analysis Tip gave an introduction to how to best explain your Monte Carlo simulation model assumption, this and the next Risk Analysis Tips will discuss how to best presenting your model results.

Two main ways

There are two main ways of describing a model's outputs:

• Graphical Descriptions (M0212)
• Statistical Descriptions (M0419)

In this Risk Analysis Tip, we will discuss the use of Graphical descriptions for a model's output. In the next Risk Analysis Tip, we will discuss the use of Statistical descriptions.

Graphical Descriptions

Monte Carlo simulation graphical outputs have the advantage over statistical outputs of providing a quick, intuitive way to understand the model's results without needing a great deal of statistical knowledge.

Often used examples of graphical descriptions include:

• Histogram and density plots (M0214) - to get the best visual understanding of the output
• Cumulative distribution plots (M0117) - to read off probabilities
• Overlaying histogram and cumulative plots (M0344) - to compare output distributions
• Time series summary plots (M0469) - to show summary effect of a time series model
• Scatter plots (M0404) - to show correlation relationships
• Tornado charts (M0474) - crude but quick sensitivity analysis plots
• Risk-return plots (M0402) - to compare different decision options by benefit and cost
• Spider plots (M0416) - advanced sensitivity analysis plots

In the section below, we will discuss the Histogram and density plot. In future Risk Analysis Tip editions we will discuss the use of the other graphical outputs.

Histogram and density Plot

The histogram, (also known as relative frequency), plot is the most commonly used in risk analysis. It is produced by grouping the data generated for a model's output into a number of bars or classes. The main advantage of a histogram plot is how easy it is to read (1) the realistic range of the variable, (2) its rough location, and (3) any peaks. We can also easily recognize common distributions like a triangular, normal, uniform, etc, and we can see whether a variable is skewed.

Overlaying histogram and cumulative plots

Several histogram plots can be overlaid. You can select several outputs and plot all histograms together. The plots are easier to read if the histograms are formatted into line plots rather than histogram bars:

Overlaying density plots like this is an intuitive way of comparing the location, spread and shape of two or more distributions. It is not very useful, however, for comparing probabilities. In addition, if one distribution has a much narrower range than the others, the plot will be very flat for the distributions with wider spread.

We therefore recommend that a complimentary cumulative distribution plot is provided alongside the histogram (density) plot to provide the maximum information.

What's next?

ModelAssist

• The material within this 'Risk Analysis Tip' comes from one of the over 500 risk analysis topics available in ModelAssist, which gives a more detailed explanation of the above methods and any risk analysis techniques involved.

top