Skip to the main content

Original scientific paper

COMBINING PCA ANALYSIS AND ARTIFICIAL NEURAL NETWORKS IN MODELLING ENTREPRENEURIAL INTENTIONS OF STUDENTS

Marijana Zekić-Sušac ; Faculty of Economics, University of J.J. Strossmayer in Osijek, Osijek, Croatia
Nataša Šarlija orcid id orcid.org/0000-0003-2600-9735 ; Faculty of Economics, University of J.J. Strossmayer in Osijek, Osijek, Croatia
Sanja Pfeifer orcid id orcid.org/0000-0002-7394-3080 ; Faculty of Economics, University of J.J. Strossmayer in Osijek, Osijek, Croatia


Full text: english pdf 161 Kb

page 306-317

downloads: 8.091

cite


Abstract

Despite increased interest in the entrepreneurial intentions and career choices of young adults, reliable prediction models are yet to be developed. Two nonparametric methods were used in this paper to
model entrepreneurial intentions: principal component analysis (PCA) and artificial neural networks (ANNs). PCA was used to perform feature extraction in the first stage of modelling, while artificial neural networks were used to classify students according to their entrepreneurial intentions in the second stage. Four modelling strategies were tested in order to find the most efficient model. Dataset
was collected in an international survey on entrepreneurship self-efficacy and identity. Variables describe students’ demographics, education, attitudes, social and cultural norms, self-efficacy and
other characteristics. The research reveals benefits from the combination of the PCA and ANNs in modeling entrepreneurial intentions, and provides some ideas for further research.

Keywords

Classification; Entrepreneurial intentions; Modelling; Artificial neural networks; Principal component analysis

Hrčak ID:

97407

URI

https://hrcak.srce.hr/97407

Publication date:

1.2.2013.

Visits: 9.434 *