Improved Model for Attribute Selection on High-Dimensional Economic Data
| Authors | |
|---|---|
| Year of publication | 2014 |
| Type | Article in Proceedings |
| Conference | Proceedings of the 2nd International Conference on Management, Leadership and Governance ICMLG 2014 |
| MU Faculty or unit | |
| Citation | |
| Field | Management and administrative |
| Keywords | Corporate Competitiveness; Financial Performance; Non-linear Regression; Feature Selection; Statistical Pattern Recognition |
| Description | This paper represents a continuation of our previous results, which were closely linked to the topic of automated search for factors of corporate competitiveness. The main goal remains to demonstrate a significant mutual dependency between corporate competitiveness (characterized mainly by their financial performance) and a group of selected characteristics describing these companies. Such characteristics can be regarded as competitiveness factors. Characteristics are generally not mutually independent, thus factors have to be selected in multidimensional space. Compared to our previous work presented at ICMLG 2013 in Bangkok, we analyse here a more precise and larger dataset of enterprises from the Czech Republic and, moreover, with a higher dimensionality (with more variables or features). This paper presents both the new improved algorithms for using the regression model in the search for key factors of corporate competitiveness and also new results achieved with this larger dataset. |
| Related projects: |