In this paper I use a unique data set from Nicaragua to asses the behavior of persons who send money back home. I estimate a heteroskedastic Tobit with a known form of variance to estimate the correlation of the remitting decisions of migrants. Working, residing in a developed country and belonging to the nuclear family positively affect remittances. The labor status and the level of education of the head of the household both affect remittances. The decision to participate in the remitting process appears to be positively related across migrants within the same receiving household.
Migration remittances censored regression decision making Central America Nicaragua