Histogram Plots

See also:

The histogram, or relative frequency, plot is the most commonly used in risk analysis. A histogram plot of simulation data can be produced in ModelRisk by selecting the variable in the Simulation Results window and clicking:

The plot is produced by grouping the data generated for a model’s output into a number of bars or classes. The number of values in any class is its frequency. The frequency divided by the total number of values gives an approximate probability that the output variable will lie in that class’ range. We can easily recognise common distributions like a triangular, normal, uniform, etc, and we can see whether a variable is skewed. The figure below shows a typical plot:

Sliders can be switched on and off by clicking . These sliders allow you to see the probability of lying within specified regions of the distribution’s range. The sliders can be moved by simply dragging then to the left and right with the mouse. The horizontal axis range of the plot can be changed by clicking , and a legend switched on and off by clicking .

When plotting two or more variables together a histogram plot can become confusing because the height of the bars of variables will be obscured by other variables plotted over them, so an alternative line plot is available by clicking  and selecting ‘Lines’.

The most common mistake in interpreting a histogram is to read off the y-scale value as the probability of the x-value occurring. In fact, the probability of any x-value, given the output is continuous (and most are), is infinitely small. If the model’s output is discrete, the histogram will show the probability of each allowable x-value, providing the class width is less than or equal to the distance between each allowable x-value.