# Heston - How important are the initial guess in calibration and if it is very important, what would be a good way to get initial guess?

So I have been trying to implement a simple Heston calibration using crude MC with 10k scenarios and 1000 time steps and the best I could get is 3x of the observed implied volatility.

I suspect it has something to do with the way my initial guess worked, and therefore, I am just wondering:

1. Is the initial guess very critical (that a not-so-great initial guess could give you 3x the differences)

2. If it is, how can I get a good initial guess?

-
3x sounds quite off, the initial guess should not get you that far away. Care to elaborate a bit on your setup. Maybe it is just a small issue. Matlab? – Matt Wolf Jun 18 '13 at 2:48
What happens if you set the initial guess (very) close to the correct answer? If that doesn't work I fear you have a bug. – Bob Jansen Jun 18 '13 at 5:35

To check your results, you might try "The Heston Model: A Practical Approach with Matlab Code" by Nimalin Moodley, http://math.nyu.edu/~atm262/fall06/compmethods/a1/nimalinmoodley.pdf , in particular the www.ingber.com open source C++ code for Adaptive Simulated Annealing (+ SWIG to wrap/parse it to the language you are using)

-
thanks a lot, this seems to be a good way to start! – AZhu Jun 19 '13 at 17:45

It depends on the used optimization algorithm, esp. whether they act locally or globally.

Just to give you some ideas:

1. Local (deterministic) algorithms (e.g. gradient methods): a good initial guess is crucial.
2. (Global) stochastic algorithms (e.g. simulated annealing): the initial guess is irrelevant.
-