Volume 20 - Article 25 | Pages 599–622  

Graduating the age-specific fertility pattern using Support Vector Machines

By Anastasia Kostaki, Javier Moguerza, Alberto Olivares, Stelios Psarakis

Abstract

A topic of interest in demographic literature is the graduation of the age-specific fertility pattern. A standard graduation technique extensively used by demographers is to fit parametric models that accurately reproduce it. Non-parametric statistical methodology might be alternatively used for this graduation purpose. Support Vector Machines (SVM) is a non-parametric methodology that could be utilized for fertility graduation purposes. This paper evaluates the SVM techniques as tools for graduating fertility rates In that we apply these techniques to empirical age specific fertility rates from a variety of populations, time period, and cohorts. Additionally, for comparison reasons we also fit known parametric models to the same empirical data sets.

Author's Affiliation

• Anastasia Kostaki - Athens University of Economics and Business, Greece EMAIL
• Javier Moguerza - Universidad Rey Juan Carlos, Spain EMAIL
• Alberto Olivares - Universidad Rey Juan Carlos, Spain EMAIL
• Stelios Psarakis - Athens University of Economics and Business, Greece EMAIL

Other articles by the same author/authors in Demographic Research

Modeling fertility in modern populations
Volume 16 - Article 6

Expanding an abridged life table
Volume 5 - Article 1

Similar article in Demographic Research

Cohort Reproductive Patterns in the Nordic Countries
Volume 5 - Article 5    | Keywords: age patterns of fertility, cohort analysis, cohort fertility, low fertility, Nordic countries, prospects for below-replacement fertility, Scandinavia