Constructing a portfolio of investments is one of the

\r\nmost significant financial decisions facing individuals and

\r\ninstitutions. In accordance with the modern portfolio theory

\r\nmaximization of return at minimal risk should be the investment goal

\r\nof any successful investor. In addition, the costs incurred when

\r\nsetting up a new portfolio or rebalancing an existing portfolio must

\r\nbe included in any realistic analysis.

\r\nIn this paper rebalancing an investment portfolio in the presence of

\r\ntransaction costs on the Croatian capital market is analyzed. The

\r\nmodel applied in the paper is an extension of the standard portfolio

\r\nmean-variance optimization model in which transaction costs are

\r\nincurred to rebalance an investment portfolio. This model allows

\r\ndifferent costs for different securities, and different costs for buying

\r\nand selling. In order to find efficient portfolio, using this model, first,

\r\nthe solution of quadratic programming problem of similar size to the

\r\nMarkowitz model, and then the solution of a linear programming

\r\nproblem have to be found. Furthermore, in the paper the impact of

\r\ntransaction costs on the efficient frontier is investigated. Moreover, it

\r\nis shown that global minimum variance portfolio on the efficient

\r\nfrontier always has the same level of the risk regardless of the amount

\r\nof transaction costs. Although efficient frontier position depends of

\r\nboth transaction costs amount and initial portfolio it can be concluded

\r\nthat extreme right portfolio on the efficient frontier always contains

\r\nonly one stock with the highest expected return and the highest risk.<\/p>\r\n","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 99, 2015"}