Decision Trees: Sensitivity Analysis, Basic and Probabilistic
Sensitivity analysis is defined as systematically varying one or more parameters in a decision model over a specified range and recalculating the [Page 350]expected utility of the model for each value. There are four reasons to employ sensitivity analysis:
- to determine the effect of reasonable variations in the estimates of parameters on the results of the analysis;
- to determine which variables are most critical to the analysis—and, therefore, may justify further efforts to estimate them more precisely;
- to determine what the analysis would recommend for various scenarios (combinations of parameters); and
- to explore the model for bugs or anomalies.
The best estimate of the value of each parameter in a model is called the baseline value. When all parameters are at their baseline values, the model ...