Investment Strategy on the Zagreb Stock Exchange Based on Dynamic DEA
Croatian Economic Survey
Nowadays, there is a growing interest in the application of quantitative methods in portfolio management, as the results of their application can be used as guidelines for managing a successful investment portfolio, i.e., a portfolio that outperforms the market. This paper deals with the Data Envelopment Analysis (DEA) approach and a Dynamic Slacks-Based Measure as a method of forming a portfolio which would predominantly outperform the market. In order to test the strategy, data on stocks
... data on stocks listed on the Zagreb Stock Exchange were gathered for the period April 2009 -June 2012. Using the quarterly returns, standard deviations and coefficients of skewness as links, a dynamic slacks-based measure approach was applied to evaluate the relative efficiency of stocks in each quarter. The findings indicate that a portfolio based on the results of the optimization beats the market in terms of both returns and risk. This is the first implementation of Tihana Škrinjarić the dynamic DEA model in stock trading. The results suggest that it is superior to basic DEA models. JEL classification: C14, C61, G11 1 Mathematical models and tools such as the Markowitz model (1952), multi-criteria modeling, Data Envelopment Analysis, statistical and econometric models and methods (regression, ARIMA models, GARCH models, VAR models, etc.). Additionally, fundamental and technical analyses are also popular in the field of stock investing. 2 Linear and non-linear programing, transportation and assignment problems, network optimization models, Markov chains methodology, many Data Envelopment Analysis models, etc.