- citati u SCIndeksu: 0
- citati u CrossRef-u:0
- citati u Google Scholaru:[
 ]
- posete u poslednjih 30 dana:1
- preuzimanja u poslednjih 30 dana:0
|
|
|
2013, vol. 6, br. 3, str. 1-17
|
|
Primena portfolio tehnika u poslovanju kompanija na finansijskoj berzi
The application of portfolio techniques to company operations in stock markets
Sažetak
Finansijsko tržište je kompleksan, nelinearan i nestacioniran dinamički sistem i kao takav zahteva napredne tehnike za njegovu analizu. Da bi se smanjio finansijski rizik, investitori moraju da izgrade portfolio kao skup investicija u različite hartije od vrednosti. U ovom radu opisuje se Markowitz-eva portfolio teorija i upoređuje sa novim pristupima koji se oslanjaju na prediktivno modelovanje cena akcija na finansijskoj berzi. Glavne slabosti Markowitz-eve portfolio teorije predstavljene su zajedno Sa predlozima za njihovo prevazilaženje. Istraživali smo mogućnosti primene tehnike rudarenja podataka (data mining) u kreiranju optimalnih portfolia. Posebno smo eksperimentisali sa neuronskim mrežama koje se učestalo koriste u procesima investicionog odlučivanja. Eksperimentalni podaci preuzeti su sa Google Finance-a i obuhvataju dnevne promene cena akcija 20 analiziranih kompanija za vremenski period od 10 godina. Dobijeni rezultati pokazuju da nova portfolio metoda zasnovana na neuronskim mrežama obećava i proizvodi visoke prinose investicija.
Abstract
Financial market is a complex, non-linear and non-stationary dynamic system and as such requires advanced analysis techniques. To reduce financial risk, investors must develop a portfolio to allocate investment into a set of different assets. In this paper we describe the Markowitz portfolio theory and compare it to a new approach that relies on predictive modeling of stock market prices. The main weaknesses of the Markowitz portfolio theory are presented along with possible suggestions for overcoming them. We also explored the possibility of incorporating data mining techniques into the creation of an optimal portfolio of securities. We especially experimented with neural networks that have been extensively used in the investment decision-making processes. Experimental data were obtained from Google Finance and they include daily changes in stock prices of 20 companies in the interval of10 years. The obtained results show that the new portfolio method based on neural networks is promising and brings a high return on investment.
|
|
|
|
Reference
|
|
|
*** (2013) Stock market quotes. Google Finance. http://www.google.com/finance
|
|
 
|
Alihodžić, A. (2010) Moderna portfolio teorija i diversifikacija. Bankarstvo, vol. 39, br. 11-12, str. 62-77
|
|
 
|
Freitas, F.D., de Souza, A.F., de Almeida, A.R. (2009) Prediction-based portfolio optimization model using neural networks. Neurocomputing, 72(10-12): 2155-2170
|
|
7
|
Markowitz, H. (1952) Portfolio selection. Journal of Finance, 7(1): 77-91
|
|
2
|
Markowitz, H.M. (1991) Foundations of Portfolio Theory. Journal of Finance, 46(2): 469-477
|
|
|
Milosavljević, M. (2005) Neuronske mreže - radni material. Beograd: Elektrotehnički fakultet
|
|
13
|
van Horne, J.C. (1995) Finansijsko upravljanje i politika. Zagreb: MATE
|
|
20
|
Vunjak, N., Ćurčić, U., Kovačević, L. (2008) Korporativno i investiciono bankarstvo. Bečej: Proleter
|
|
2
|
Williams, J.B. (1938) The theory of investment value. Cambridge, MA, itd: Harvard University Press, str. 55-75
|
|
|
|
|