Search code examples
pythonoptimizationcalibrationagent-based-modelingstochastic

Calibration of a stochastic python simulation


I have an (agent-based) simulation in python. The results are stochastic. I would like to calibrate the simulation be minimizing a sum of square distances. What python algorithms are available to minimize a stochastic simulation?


Solution

  • Without knowing more, for example how is "distance" calculated? I can only give general advice. I suggest scipy's optimization module. Depending on whether the sum is a scalar or a vector, you can use a different optimizer.

    If you are unfamiliar with the different optimization methods, I suggest looking through this brief overview.