DEA Bootstrap

DEA Bootstrap dilakukan melalui dua prosedur, yaitu menghitung skor efisiensi terlebih dahulu, kemudian mempergunakan analisis regresi untuk menjelaskan keragaman daripada skor-skor efisiensi tersebut. Regresi Ordinary Least Square (OLS) memiliki keterbatasan dalam analisa keragaman skor efisiensi DEA, dikarenakan skor DEA tersebut sangat berhubungan (berkorelasi) erat dengan variabel bebas pembentuknya (pada proses perhitungan skor DEA pada tahapan analisa data), sehingga nilai estimasi regresi dapat bias (Simar, 1992).

Di sisi lain, terdapat beberapa pendekatan untuk menyelesaikan permasalahan pendugaan keragaman skor efisiensi DEA dengan regresi (Xue dan Harker, 1999; Casu dan Molineux, 1999). Pendekatan ini dilakukan oleh Xue dan Harker (1999): menitikberatkan bahwa skor efisiensi yang dihasilkan model DEA jelas bergantung

sama lain dalam analisis statistik.

Alasan dependensi ini sebenarnya merupakan fakta yang umum diketahui bahwa skor efisiensi DEA sendiri adalah indeks relatif efisiensi, bukan indeks efisiensi absolut. Dikarenakan keberadaan dependensi inheren di antara skor efisiensi, salah satu asumsi analisis regresi konvensional, independensi di dalam sampel (autokorelasi), dilanggar. Sehingga, prosedur regresi konvensional (uji asumsi klasik) menjadi tidak valid. Untuk langkah alternatifnya, Xue dan Harker (1999) serta Casu dan Molineux (1999) melakukan regresi bootstrap.

Kerangka "COOPER" dalam DEA

In large and complicated datasets, a standard process could facilitate performance assessment and help to (1) translate the aim of the performance measurement to a series of small tasks, (2) select homogeneous DMUs and suggest an appropriate input/output selection, (3) detect a suitable model, (4) provide means for evaluating the effectiveness of the results, and (5) suggest a proper solution to improve the efficiency and productivity of entities (also called Decision Making Units, DMUs). 

We suggest a framework which involves six interrelated phases: (1) Concepts and objectives, (2) On structuring data, (3) Operational models, (4) a Performance comparison model, (5) Evaluation, and (6) Results and deployment. Taking the first letter of each phase, we obtain the COOPER-framework (in honour of and in agreement with one of the founders of DEA). Figure 1 systemizes the six phases.

BUKU: Islamic Banking Efficiency: Efficiency Of Islamic Banks In Pakistan using Data Envelopment Analysis

Islamic banking is one of the most growing sectors of financial market and gaining popularity in Islamic world. With increasing competition and advances in banking systems Islamic banks must be efficient to reap the benefits of growing demand. 

This book investigates the efficiency of Islamic banks in Pakistan using non-parametric approach of Data Envelopment Analysis (DEA). The purpose is to look at the financial characteristics that make Islamic banks efficient. Keep in view the financial characteristics of performance, current study apart efficient Islamic banks from those that are found inefficient. 

The efficiency of Islamic banks is measured in specified input and output variables. Staff cost, fixed assets and total deposits are taken as input variables while total loans, income and liquid assets are taken as output variables.