Proxy modeling
Proxy model sometimes has several meanings but most usually refers as a mathematically or statistically defined function that replicates the simulation model output for selected input parameters.[1]
Why and where to use proxy modeling
Proxy model used when simulation run takes too long and to perform many simulations is inappropriate.
| + | - | |
| Monte Carlo | Simulation results | Requires many simulation runs |
| Proxy Model | Requires less simulation runs | Approximate results |
Experimental Design
Experimental design - is a combination of the input parameter values such that maximum information can be obtained from minimum number of simulation runs. One of such designs is orthogonal array design.
Minimum number for runs:[2]
A response model is a polynomial regression model. Ususally a 2nd order regression used, for multiple parameters x takes the general form:
See also
External links
Online Multiple Polynomial Regression