A macro-economic model to forecast remittances based on Monte-Carlo simulation and artificial intelligence

A computer system based on Monte-Carlo simulation and fuzzy logic has been designed, developed and tested to: (i) identify covariates that influence remittances received in a specific country and (ii) explain their behavior throughout the time span involved. The resulting remittance model was designed theoretically, identifying the variables which determined remittances and their dependence relationships, and then developed into a computer cluster. This model aims to be global and is useful for assessing the long term evolution of remittances in scenarios where a rich country is the host (United States of America) while a poor country is the where the migrant is from (El Salvador). By changing the socio-economic characteristics of the countries involved, experts can analyze new socio-economic frameworks to obtain useful conclusions for decision-making processes involving development and sustainability.

► Bayesian network models where the relationships between variables are fuzzy.
► Forecasting in time-dependent Bayesian networks using simulation and fuzzy logic.
► Uncertainty modeling when the number of rules in the rule base is huge.
► Remittance assessment in unstable and unbalance economic frameworks.