Application of regression models and artificial neural networks in agriculture –Prediction in spirit drinks sector (CROSBI ID 202286)
Prilog u časopisu | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Čačić, Jasna ; Gajdoš Kljusurić, Jasenka ; Čačić, Dražen
engleski
Application of regression models and artificial neural networks in agriculture –Prediction in spirit drinks sector
Croatian accession to the European Union (EU) was one of the main goals of Croatia in the last decade. However, after Croatia accedes to the EU new challenges and opportunities for agriculture and food industry will arise. The implications of the accession on the Croatian spirit drinks sector could be diverse and may cause production and market difficulties. Reliable prediction models for the spirit drinks sector are not available ; thus, this research aimed to (i) analyse current situation in the Croatian spirit drinks sector and (ii) to define changes that might occur within the sector as a result of Croatian accession to the EU as well as to (iii) predict sector’s trends in export, import and consumption per capita. In order to predict trend changes in the spirit drinks sector the following research methods were used: SWOT analysis, regression models, principal component analysis (PCA) and artificial neural networks. Research models explored the influence on the production, export and import. Models were shaped on the basis of the “training” (model) countries for which were taken EU members that are similar to Croatia in their historical, political and social background. Effectiveness of the prediction models was confirmed for the year 2011. Thus, the prediction models should be used when adjusting expectations and actions for the spirit drinks sector to either positive or negative trends.
Croatia; modelling; spirit drinks sector; SWOT; PCA; neural networks
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano