help dea (SJ10-2: st0193)
---------------------------------------------------------------------------------------------------------------------------
Title
dea -- Data envelopment analysis
Syntax
dea ivars = ovars [if] [in] [, options]
options Description
---------------------------------------------------------------------------------------------------------------------
rts(crs|vrs|drs|nirs) specify the returns to scale; default is rts(crs)
ort(in|out) specify the orientation; default is ort(in)
stage(1|2) specify the way to identify all efficiency slacks; default is stage(2)
trace save all sequences and results from Results window to dea.log
saving(filename) save results to filename; save statistics in double precision; save results to filename every
# replications
---------------------------------------------------------------------------------------------------------------------
Description
The dea command requires an initial dataset that contains the input and output variables for observed units. dea
selects the input and output variables from the user-designated data file or the dataset currently in memory and
solves data envelopment analysis (DEA) models with the options specified.
Variable names must be identified in ivars for input variables and in ovars for output variables so that dea can
identify and handle the multiple input-output dataset. In the output of the dea command, the prefix dmu: precedes
decision-making unit (DMU) names.
dea has the ability to accommodate an unlimited number of inputs and outputs with an unlimited number of DMUs. The
only limitation is the available computer memory.
The resulting file reports information including reference points and slacks in DEA models. This information can be
used to analyze the inefficient units, for example, the source of the inefficiency and how an inefficient unit could
be improved to the desired level.
The saving(filename) option creates a filename.dta file that contains the results of dea, including information about
the DMU, input and output the data used, ranks of DMUs, efficiency scores, reference sets, and slacks.
The log file, dea.log, will be created in the working directory.
Options
rts(crs|vrs|drs|nirs) specifies the returns to scale. The default, rts(crs), specifies constant returns to scale.
rts(vrs), rts(drs), and rts(nirs) specify variable returns to scale, decreasing returns to scale, and
nonincreasing returns to scale, respectively.
ort(in|out) specifies the orientation. The default is ort(in), meaning input-oriented DEA. ort(out) is
output-oriented DEA.
stage(1|2) specifies the way to identify all efficiency slacks. The default is stage(2), meaning two-stage DEA.
stage(1) is single-stage DEA.
trace specifies that all the sequences displayed in the Results window also be saved in the dea.log file. The
default is to save the final results in the dea.log file.
saving(filename) specifies that the results be saved in filename.dta. If filename.dta already exists, the existing
data will be moved to the file filename_bak_DMYhms.dta before the new data are saved in filename.dta.
Examples
. use coelli_table6.4.dta
. dea i_x = o_q
. dea i_x = o_q, rts(vrs)
. dea i_x = o_q, rts(vrs) ort(out)
. dea i_x = o_q, rts(vrs) ort(out) stage(2)
. dea i_x = o_q, rts(vrs) ort(out) stage(2) saving(dea1_result)
Saved results
dea saves the following in r():
Matrices
r(dearslt) results of dea that have observation rows of DMUs and variable columns with input data, output data,
efficiency scores, references, slacks, and more depending on the model specified.
Authors
Yong-bae Ji
Korea National Defense University
Seoul, Republic of Korea
Choonjoo Lee
Korea National Defense University
Seoul, Republic of Korea
sarang90@kndu.ac.kr;
sarang64@snu.ac.kr;
bloom.rambike@gmail.com
Also see
Article: Stata Journal, volume 10, number 2: st0193
扫码加好友，拉您进群
全部版块
我的主页
收藏
