20th International Symposium on Econometrics, Operations Research and Statistics, Ankara, Turkey, 12 February 2020, pp.163
Borsa Istanbul A.Ş. (BIST) has the city indices that allow investors to invest in a particular city. Several studies analyze return performances of the city indices from BIST. However, efficient results are obtained with the help of analyzes enriched with new approaches. In this analysis we aim to examine return performances of city indices from BIST by using Dynamic Data Envelopment Analysis which take into consideration an intermediate input/output variable to link consecutive periods in the time interval.
Data Envelopment Analysis (DEA) is a linear programming based technique which is used to measure the relative efficiencies of Decision Making Units (DMUs). DMUs use same kind of inputs to produce same kind of outputs, in other words they should be homogenous. Traditional DEA technique consider the DMUs for a specific time or process via cross-section data, then the relative efficiency scores are obtained for a separate time.
In this study, we focus on Dynamic DEA to measure the relative efficiency of the return of the City Indices in BIST. In our analysis, we try to measure the efficiency of city indices regarding their weekly returns. The relative efficiency scores of city indices are evaluated for the time interval between 2015-2018 in Dynamic DEA. Input of the model is considered as absolute value of correlation coefficients of index return series and the output of the model is mean gross return over standard deviation of returns. Additionally, to ensure the dynamic procedure, skewness of return of city index is used as an intermediate input/output variable to link consecutive periods in the time interval. One period’s output is used as an input for the next time period then the intermediate input/output helps to investigate DMUs time period instead of a specific time. With this feature, the analysis is an example of usage a panel data.
According to the results, Antalya and Tekirdağ city indices are not efficient among the other 10 indices. The used Dynamic DEA model labeled a DMU as efficient if it is efficient at least one term when the DMU is considered via traditional static DEA technique. Since, Antalya and Tekirdağ city indices are not efficient for any time overall efficiency scores according to Dynamic DEA is less than 1. Although the dynamic efficiency scores are equal to 1, there is no city indices that is efficient for all the periods separately. From this perspective static DEA and dynamic DEA can be used together. Dynamic DEA is a powerful tool to link the consecutive time periods, on the other hand static DEA provides to handle separate terms individually.
From the perspective of our analysis, a city that is efficient for all periods makes more sense in terms of investment, there is no such index within the time period examined. The results are important for researchers and investors seeking profitable investment tools and also can be evaluated in terms of regionally in future works. Even though our analysis based on a specific period of time and the data, the implications arising from the results of the study are important for the companies’ policies. As a result, when static and dynamic DEA are evaluated together, it helps to make strong decisions. They have no superior to each other.