In the previous article, “Using Data Envelopment Analysis (DEA) program for measuring efficiency” the process to download Data Envelopment Analysis program (DEAP) was explained and the case problem in hand was defined. The first step is to understand how to import data to DEAP. MS Excel is used to compile the data collected from the various sources and literature followed in selection of input and output variables. The underlying steps will help in understanding importing and importance of each part.
Importing data from MS Excel
First step is to copy the variables from MS Excel to Notepad (banks.txt akin to data.txt). Change the file name accordingly.
Quick Note: First add the Output variables (i.e. Total loans and Profits) and then add the Input variables (i.e. Total Capital and Deposits) for the 10 case banks while importing the data to Data Envelopment Analysis (DEA).
Defining the variables and files names in DEA
Then open the instruction file, data-ins.txt as in the DEA tool package (file name unchangeable as found while practical application). As in the above image banks.txt is file location for data. Give any name to output file. Here,
- Decision making units (DMUs’) are defined by no. of firms or no. of rows. 10 in the current case example or 10 DMUs’.
- Mention the output file name where the dataset pasted (banks.txt) and also mention the name of the file where the results represented (result.txt).
- Number of time period, here is “one” because performance efficiency for only one year is being analyzed (FY2016).
- No. of outputs and inputs are 2 as mentioned previously wherein Inputs are Total Capital and Deposits and Outputs are Total Loans and Profits.
- This software supports both constant returns to scale (CRS) and variable returns to scale (VRS) scales of analyzing DEA. However, one may select either 0 for CRS and 1 for VRS according to the personal requirement. This instruction file is unchangeable, any unprecedented change may cause wrong results. Here CRS based analysis used and hence “0” is chosen. In the similar way for an input oriented CRS- DEA; where the efficiency testing based on the level of input variables. Choose “0” or if it is output oriented then choose “1”.
- Again, there are different methods to perform a data envelopment analysis which has been classified as Multistage, Cost-DEA, MALMQUIST, 1-Stage DEA and 2-Stage DEA. “Multistage” model used in most cases to perform efficiency analysis, due to Radial and Slack movement analytics to provide projected values of efficiency measure and so choose “0” in the instruction file. Hence, multi stage analysis has been adopted in this study.
Output Results of different models
|Cost DEA||Technical efficiency, Allocative efficiency and Cost efficiency, Summary of cost minimizing values (input/output)|
|MALMQUIST||Distance summary, Technical Efficiency, MALMQUIST Index Summary, Index Summary (Annual means and Firm means)|
|1-Stage DEA||Same as multi-stage but Algorithm is different (linear programming with Slacks residuals)|
|2-Stage DEA||Same as multi-stage but Algorithm is different (linear programming)|
|Multistage||Efficiency summary, Slacks, Peers, Target summary, Technical Efficiency and Firm wise results|
Different types of models and their outputs
Next step is to open DEAP.EXE and type the instruction file name (e.g. data-ins.txt) and run. This is also the basic interface of the DEA tool which is a DOS command file system. The file name data-ins.txt and press enter. The analysis will hardly take 10 secs.
Running the DEA software
The instruction file has an output file named results.txt. So, in the folder named DEAP, look for the output file that will be in .txt format and open it. The results file or the output file will shows the efficiency and other related results.
Here one should note that these instructions are basically to perform input oriented multistage model based CRS-DEA. The findings and interpretations from the analysis discussed and represented in the next article.
- Coelli, T. J. (2008). A Guide to DEAP Version 2.1: A Data Envelopment Analysis (Computer) Program. CEPA Working Papers, 1–50. Retrieved from https://absalon.itslearning.com/data/ku/103018/publications/coelli96.pdf.
Latest posts by Avishek Majumder (see all)
- Importance of machine learning in biomarker research - January 22, 2019
- Gaps in health insurance plans are resulting in increasing allergy prevalence in India - January 19, 2019
- Impact of international policies on the Indian Indian drug market - January 19, 2019