Türk yenilenebilir enerji sektörünün analizi: Bir bootstrap-veri zarflama analizi yaklaşımı

---

Thesis Type: Postgraduate

Institution Of The Thesis: Gazi University, Fen Bilimleri Enstitüsü, İSTATİSTİK ANABİLİM DALI, Turkey

Approval Date: 2017

Thesis Language: Turkish

Student: ESRA BETÜL KINACI

Supervisor: HACI HASAN ÖRKCÜ

Open Archive Collection: AVESIS Open Access Collection

Abstract:

Data Envelopment Analysis (DEA) is a linear programming based method and is used to evaluate the relative efficiency of Decision-Making Units (DMU) responsible for output or outputs using similar inputs. DEA is one of the most preferred methods since it does not want a functional relation between input and outputs. Estimation of the efficiency scores obtained with DEA is a bias. This is one of the disadvantages of DEA. Especially in small-sized observations, this bias will get bigger values. To solve this problem, Lönthgren and Tambour (1997) and Simar and Wilson (1997,2000) combined the VZA method with a bootstrap method and proposed a new approach. In this way, the efficiency scores can be estimated. Later on, the efficiency scores were purged from of bias value and unbiased efficiency scores were estimated. In this thesis study, the energy performance of Turkey in terms of consumption of renewable energy sources was evaluated with the DEA. This analysis was applied to several types of renewable energy sources and Turkey's renewable energy sector analysis was evaluated for 2013 and 2014. DMUs are composed of 21 European countries including Turkey and Russia. In the study, the experimental distribution of the efficiency scores obtained by the DEA was obtained and the bias of the efficiency scores were estimated through this experimental distribution by using different algorithm types of bootstrap method. Confidence intervals for unbiased estimates are calculated. The obtained results are given in charts and tables