Resumo

MORALES-ESLAVA, Ángel Jair; SILVA-URRUTIA, Eliud  e  ORDORICA-MELLADO, Manuel. Mexican fertility forecasts: a multivariate time series model application. Pap. poblac [online]. 2016, vol.22, n.89, pp.99-131. ISSN 2448-7147.

The steady fertility decline in Mexico is relevant for the study of the demographic dynamics of the country and for population projections. Information from the official data (registered and available on line), show that fertility pattern based on the age of the mother have been presenting different structures. The aim of this paper is to estimate forecasts through the application of a VAR model. By using this model it's possible to: a) describe the monthly evolution of the births in Mexico, and b) fit a multivariate statistical model in order to forecast this series to 2020. The data used are the registered births according by age of the mother (January 1992-December 2010). Results show that these models provide a set of very precise forecasts.

Palavras-chave : fertility; forecast; VAR model; multivariate statistical model.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )