Home // ALLDATA 2016, The Second International Conference on Big Data, Small Data, Linked Data and Open Data // View article


Analyzing the Retirement Satisfaction Predictors among Men and Women Using a Multi-Layer Feed Forward Neural Network and Decision Trees

Authors:
Ehsan Ardjmand
Gary Weckman
Diana Schwerha
Andrew Snow

Keywords: Retirement Satisfaction; Artificial Neural Networks; Multi-Layer Perceptron; Decision Tree

Abstract:
In this article, we will analyze the effect of different retirement satisfaction predictors on each other and the retirement satisfaction level among men and women. The following factors will be used as predicators of retirement satisfaction: health; wealth; smoking and drinking habits; education; faith; income; impact of health on activities of daily living (ADL); frequency of activities; and the number of people in a household. A set of 858 retired men and 1179 retired women from a 2012 Health and Retirement Study database have been chosen and analyzed. A neural network was trained for each gender in order to predict retirement satisfaction; it also generated a decision tree that symbolizes the retirement satisfaction and its predictors. The results demonstrate that health, age, smoking habits, income, and wealth are the most significant predictors for both genders, while for men, education also plays an important role in retirement satisfaction.

Pages: 102 to 107

Copyright: Copyright (c) IARIA, 2016

Publication date: February 21, 2016

Published in: conference

ISSN: 2519-8386

ISBN: 978-1-61208-457-2

Location: Lisbon, Portugal

Dates: from February 21, 2016 to February 25, 2016