Enhance maximum likelihood maximization component

[nextpagesoft]

Magda:

The likelihood maximization component – at the moment this relies (Ard correct me if I’m wrong) on simplex optimisation algorithm. Following that a bootstrap procedure is used, generating new possible datasets for which the parameters are estimated. Potentially a different algorithm could be used. And/or a different likelihood component if we think of Bayesian structure with some priors.