4. Model
simulation to reproduce behavior of a
system
Enables prediction (predictive)
Enables control of a system (perscriptive)
5. Model Validation
validation is necessary for both predictive
and prescriptive model use.
Validation exercises
improve model performance
Improve insight.
6. Basic Model
Maximize f(X) - g(Z)
s.t. X -DY ≤ 0
GY - Z ≤ 0
AY ≤ b
X,Y,Z ≥ 0
The optimal values, X*, Y*, Z are assumed to correspond to real world
observations X’, Y’, Z’. The model also has shadow prices U*, V*, W*
that correspond to real world values U’, V’, W’.
X is some output measure. Z is some input measures.
13. Validation by Results
1. Gather inputs and outputs
o 2. Use inputs on model
o 3. Gather model outputs
o 4. Compare model outputs with real-world outputs
o 5. Determine validity
15. Prediction Experiment
The prediction experiment is the most common validation by results test.
the ultimate validation experiment in that it tests whether a model can
replicate reality.
16. Change Experiment
To test a model's ability to predict change, one must have
data on two real world situations and the resultant model
solutions.
Then, a comparison is made between the change in the
model solution variables (e.g.,X1 *, X2 *) and the change
observed in the real world solution X1’,X2’.
17. Tracking Experiment
One may wish to track adjustments through time.
comparisons are made between changes in the model
solution and observed changes in the real world
solution
18. Comments
Validation is important
A good model is validated by construction
and by results
Ultimate test of validity is adoption of the
model by decision makers