Choosing a sampling distribution for a Monte Carlo integration to take advantage of available efficiencies.
When completing a Monte Carlo integration it makes a difference how much of the search space is occupied by the area under the curve to be integrated. The more the curve fills the space the fewer iterations are needed to complete the integration to a given accuracy.
Exploring Monte Carlo convergence issues allows us to make the trade-off between estimation error and computing resources. This can have a big impact on trading system development times.