Predviđanje ishoda različitih pojava oduvek je predstavljalo atraktivnu istraživačku temu u velikom broju naučnih disciplina, naročito u ekonomiji. Kao posebna oblast, ekonometrija pruža različite modele za predviđanje pokazatelja kao što su BDP, stopa inflacije, kamatna stopa, cena raznih roba i usluga, kao i mnogih drugih kako na mikro tako i na makro nivou. Razvoj informacionih tehnologija omogućio je obavljanje računskih operacija mnogo brže i pouzdanije. Međutim, poseban doprinos se ogleda kroz primenu rudarenja podataka u svrhu ekstrahovanja relevantnih informacija iz velikog skupa podataka. Modeli koji se razvijaju primenom rudarenja podataka pružaju dobre rezultate u predviđanju ekonomskih pokazatelja, često i uspešnije u odnosu na određene ekonometrijske modele. U ovom radu nastoji se predvideti promena (rast) BDP-a kroz primenu rudarenja vremenskih serija na primeru Republike Srbije. Analiza je obavljena u dva slučaja: u jednom modeli uključuju nezavisne atribute koji dodatno opisuju zavisnu promenljivu, dok drugi slučaj ne sadrži ove atribute. Upotrebljene su tri različite metode rudarenja u oba slučaja (linearna regresija, višeslojni perceptron i metoda slučajne šume), a dobijeni rezultati uspešnosti modela su predstavljeni i protumačeni.
Breiman L. Random Forest. Machine learning. 2001;5–32.
5.
Carreiro A, Clark T, Marcellino M. Bayesian vars: specification choices and forecast accuracy. Jurnal of applied econometrics. 2015;46–73.
6.
Carreiro A, Clark T, Marcellino M. Large Bayesian vector autoregressions with stochastic volatility and non-conjugate priors. Journal of Econometrics. 2019;137–54.
7.
Carriero A, Galvao A, Kapetanios G. A comprehensive evaluation of macroeconomic forecasting methods. International Journal of Forecasting. 2019;1226–39.
8.
Chatfield C. 2016;
9.
Dean J. Data mining and machine learning, Value Creation for Business Leaders and Practitioners. 2014;
10.
Erkam G, Kayakutlu G, Daim T. Using artificial neural network models in stock market index prediction. Expert Systems with Applications. 2011;10389–97.
11.
Frank E, Hall M, Witten I, Pal C. The WEKA Workbench. 2016;
12.
Gérard B, Scornet E. A random forest guided tour. 2016;197–227.
13.
Hall M. Time Series Analysis and Forecasting with Weka -Pentaho Data Mining. 2014;
14.
Kriesel D. A Brief Introduction to Neural Networks. 2007;
15.
Limited F. 2022;
16.
Menzies T, Kocagüneli E, Peters F, Turhan B. 2015;321–53.
17.
Nemes M, Butoi A. Data Mining on Romanian Stock Market Using Neural Networks for Price. Informatica Economică. 2013;125–36.
18.
Schorfheide F, Song D. Real-Time Forecasting with a Mixed-Frequency VAR. Journal of Business & Economic Statistics. 2014;
19.
Shmueli G, Bruce C, Patel R. 2016;
20.
Smets F, Warne A, Wouters R. Professional forecasters and real-time forecasting with a DSGE model. International Journal of Forecasting. 2014;981–95.
21.
R. Republički zavod za statistiku. 2022;
22.
Stock H, Watson W. 2015;
23.
Tsai CF, Wu JW. Using neural network ensembles for bankruptcy prediction and credit scoring. Expert Systems with Applications. 2008;2639–49.
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.
