Download
a complete copy of this risk analysis resource for free here.
|
Targets are the variables that are to be optimized. For example, one might wish to minimize the expected (mean) cost of a project or number of people made redundant; or maximize the expected profit of a venture; or minimize the standard deviation (spread) of the performance specification of a machine; or minimize the 99th percentile of potential trading losses.
Clicking the Add button in the Targets tab opens the following dialog:
Location is the cell position of the variable to be optimized
Type provides a list of options: minimize; maximize; or set to a specific value (Value). If one chooses ‘Value’ the Value field is enabled to enter the desired target value
Statistic allows one to specify a statistic for the variable that is to be optimized. In the dialog above the mean value of the target is to be minimized. Options are: mean; median; a percentile (in which case the Percentile value field becomes active requiring an input on (0,1)); min; max; stdev (standard deviation); variance; range (the difference between the minimum and maximum observed values); skewness; kurtosis; CofV (coefficient of variation); and FinValue (the final value of a simulation, used if one builds a model that simulates across iterations in some way).
Name is a text field that identifies the target
Enabled switches the target definition on and off: ticked (=TRUE in the Excel function) indicates it is active, not ticked (FALSE) indicated that it is disabled. This allows one to have several targets defined in the model and build controls that switch those targets on and off.
Entering the information as shown above and selecting OK will add this target definition to the Optimization Settings:
In Cell B6 of the model the following ModelRisk function has now been added:
= [formula] + VoseOptTargetMaximize("Cost","Mean",TRUE)
where [formula] is the original Excel formula that was in this cell.
ModelRisk incorporates three functions for defining target cells to optimize:
VoseOptTargetMaximize(Name, Statistic, Enabled) for a variable to maximize
VoseOptTargetMinimize(Name, Statistic, Enabled) for a variable to minimize
VoseOptTargetValue(Name, Statistic, Value, Enabled) for a variable to set as close as possible to a specific Value
Options for the function parameters are:
Name: a text string describing the variable to be optimized
Statistic: “Mean”; “Median”; a value between zero and one for the Percentile option, “Min”, “Max”, “StDev”, “Variance”, “Range”, “Skewness”, Kurtosis”, and “FinValue”
Value: any real value (VoseOptTargetValue only)
Enabled: a Boolean parameter taking either TRUE (or equivalently 1) when the target definition is to be used, or FALSE (or equivalently 0) when the target definition is not to be used.
Example model