After its introduction in 1978, Data Envelopment Analysis (DEA) has instantly been recognized as a useful methodology for measuring the relative efficiency of different entities, called Decision Making Units (DMUs), given multiple criteria. Up until nowadays, the popularity of DEA has been growing and a significant number of bibliographical items was published, reporting on both theoretical and empirical results. However, the main applicative area of DEA remained the performance measurement in economics and business. On the 40th anniversary of DEA, the aim of this paper is to present the DEA bibliography of Croatian scientists (up until June 2018). We consider six main categories of DEA-related publications, followed with key statistics and an overview of keywords and research areas. The whole list of DEA-related publications used in this analysis is published online. We believe this research will shed light on the state of DEA in Croatian science and motivate future researches.
ozana nadoveza jelić, margareta gardijan kedžo: efficiency vs effectiveness: an analysis of tertiary education across europe public sector economics 42 (4) 381-414 (2018)
Recent research in the field of investor preference has emphasised the need to go beyond just simply analyzing the first two moments of a portfolio return distribution used in a MV (mean-variance) paradigm. The suggestion is to observe an investor's utility function as an n th order Taylor approximation. In such terms, the assumption is that investors prefer greater values of odd and smaller values of even moments. In order to investigate the preferences of Croatian investment funds, an analysis of the moments of their return distribution is conducted. The sample contains data on monthly returns of 30 investment funds in Croatia for the period from January 1999 to May 2014. Using the theoretical utility functions (DARA, CARA, CRRA), we compare changes in their preferences when higher moments are included. Moreover, we investigate an extension of the CAPM model in order to find out whether including higher moments can explain better the relationship between the awards and risk premium, and whether we can apply these findings to estimate preferences of Croatian institutional investors. The results indicate that Croatian institutional investors do not seek compensation for bearing greater market risk.
Abstract. Most strategies for selection portfolios focus on utilizing solely market data and implicitly assume that stock markets communicate all relevant information to all market stakeholders, and that these markets cannot be influenced by investor activities. However convenient, this is a limited approach, especially when applied to small and illiquid markets such as the Croatian market, where such assumptions are hardly realistic. Thus, there is a demand for including other sources of data, such as financial reports. Research poses the question of whether financial ratios as criteria for stock selection are of any use to Croatian investors. Financial and market data from selected publicly companies listed on the Croatian capital market are used. A two-stage portfolio selection strategy is applied, where the first stage involves selecting stocks based on the respective Data Envelopment Analysis (DEA) efficiency scores. DEA models are becoming popular in stock portfolio selection given that the methodology includes numerous models that provide a great flexibility in selecting inputs and outputs, which in turn are considered as criteria for portfolio selection. Accordingly, there is much room for improvement of the current proposed strategies for selecting portfolios. In the second stage, two portfolio-weighting strategies are applied using equal proportions and scoreweighting. To show whether these strategies create outstanding out-of-sample portfolios in time, time-dependent DEA Window Analysis is applied using a reference time of one year, and portfolio returns are compared with the market portfolio for each period. It is found that the financial data are a significant indicator of the future performance of a stock and a DEA-based portfolio strategy outperforms market return.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.