hrcak mascot   Srce   HID

Original scientific paper

SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS

Marijana Zekić-Sušac ; Faculty of Economics, University of J.J. Strossmayer in Osijek, Croatia
Nataša Šarlija   ORCID icon orcid.org/0000-0003-2600-9735 ; Faculty of Economics, University of J.J. Strossmayer in Osijek, Croatia
Mirta Benšić   ORCID icon orcid.org/0000-0001-9063-0310 ; Department of Mathematics, University of J.J. Strossmayer in Osijek, Croatia

Fulltext: english, pdf (150 KB) pages 83-95 downloads: 2.177* cite
APA 6th Edition
Zekić-Sušac, M., Šarlija, N. & Benšić, M. (2005). SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS. Journal of Information and Organizational Sciences, 29 (2), 83-95. Retrieved from https://hrcak.srce.hr/78281
MLA 8th Edition
Zekić-Sušac, Marijana, et al. "SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS." Journal of Information and Organizational Sciences, vol. 29, no. 2, 2005, pp. 83-95. https://hrcak.srce.hr/78281. Accessed 2 Apr. 2020.
Chicago 17th Edition
Zekić-Sušac, Marijana, Nataša Šarlija and Mirta Benšić. "SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS." Journal of Information and Organizational Sciences 29, no. 2 (2005): 83-95. https://hrcak.srce.hr/78281
Harvard
Zekić-Sušac, M., Šarlija, N., and Benšić, M. (2005). 'SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS', Journal of Information and Organizational Sciences, 29(2), pp. 83-95. Available at: https://hrcak.srce.hr/78281 (Accessed 02 April 2020)
Vancouver
Zekić-Sušac M, Šarlija N, Benšić M. SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS. Journal of Information and Organizational Sciences [Internet]. 2005 [cited 2020 April 02];29(2):83-95. Available from: https://hrcak.srce.hr/78281
IEEE
M. Zekić-Sušac, N. Šarlija and M. Benšić, "SELECTING NEURAL NETWORK ARCHITECTURE FOR INVESTMENT PROFITABILITY PREDICTIONS", Journal of Information and Organizational Sciences, vol.29, no. 2, pp. 83-95, 2005. [Online]. Available: https://hrcak.srce.hr/78281. [Accessed: 02 April 2020]

Abstracts
After production and operations, finance and investments are one of the most frequent areas of neural network applications in business. The lack of standardized paradigms that can determine the efficiency of certain NN architectures in a particular problem domain is still present. The selection of NN architecture needs to take into consideration the type of the problem, the nature of the data in the model, as well as some strategies based on result comparison. The paper describes previous research in that area and suggests a forward strategy for selecting best NN algorithm and structure. Since the strategy includes both parameter-based and variable-based testings, it can be used for selecting NN architectures as well as for extracting models. The backpropagation, radialbasis, modular, LVQ and probabilistic neural network algorithms were used on two independent sets: stock market and credit scoring data. The results show that neural networks give better accuracy comparing to multiple regression and logistic regression models. Since it is model-independant, the strategy can be used by researchers and professionals in other areas of application.

Keywords
neural networks; non-linear forward strategy; stock market prediction; credit scoring

Hrčak ID: 78281

URI
https://hrcak.srce.hr/78281

Visits: 2.321 *