Time series plots

Download a pdf copy of this help file  here

A spreadsheet model will often have one or more arrays of cells that are modeling the random or uncertain nature of a variable over time, for example:

• Exchange rates

• Sales market

• Volume of imports

• Price of commodities

• Growth of a bacterial population

• Oil price

ModelRisk provides the ability to produce time series plots of input or output arrays within your model. This involves naming a group of cells as a collective input or output – click here for more details on how.

Time series plots can be produced in ModelRisk by selecting a time series variable in the Simulation Results window and clicking:

In the graph above, the variable ‘Sales Volume’ is plotted from 2010 to 2015. The red central line shows the mean for each period. The light blue region shows the 25%-75% range, whilst the dark blue region shows the 1%-99% range. The plotted percentiles can be changed by clicking on .

Time series plots allow the user to quickly appreciate the behavior of the trend and spread in the variable over time, as well as more subtle notions of cyclical or periodic behavior.

Note: Competing Monte Carlo software usually offer the option of plotting bounds around the mean in terms of standard deviations. This option is not available in ModelRisk because a spread of say 1 standard deviation around the mean will encompass a varying percentage of the distribution depending on its form. That means that there is no consistent probability interpretation attached to mean +/- x standard deviations, and such graphs are often misinterpreted. If you feel that your variables are roughly normally, the following percentile ranges will give standard deviation spreads:

Standard deviations Low percentile High percentile ranges will give standard deviation spreads: