Selecting occupations to create jobs using discrete choice models: Medellin Metropolitan Area 2009

The main objective of this paper is to identify key factors that determine the decision of an individual to become an entrepreneur in the Medellin Metropolitan Area in 2009. Using models such as Multinomial Logit and Probit, as well as Sequential Binary Logit, we assess the probability that a person makes an occupational decision to be an entrepreneur, independent formal, informal or employee, the first state being the one that has the greatest effect on regional development. The results show us that education, use of information technologies and telecommunications, and financial capital are the variables that contribute the most to the probability of being an entrepreneur. These results may add new elements to the public policy discussion related to entrepreneurship and company growth.