About "Teradial" code

[Gizem KAYA]

Hello,

I am working on a study which is about airlines' efficiency by using "Teradial" STATA ado file.

I had a problem with the analyze.
Even though I have installed the ado file, the STATA gives the following error:

Radial (Debreu-Farrell) output-based measures of technical efficiency under assumption of CRS
technology are computed for the following data:

Number of data points (K) = 69
Number of outputs (M) = 1
Number of inputs (N) = 4

Data for reference set are not provided. Reference set is formed by 69 data points, for which
measures of technical efficiency are computed
unrecognized command
r(199);


Could you help me with this error?

Thanks in advance.

[Clyde Schechter]

Well, I don't know anything about this particular command. But clearly it begins to run and then breaks over some unrecognized command. But after finding it, I see that it is part of a package that includes a number of other .ado files. Perhaps -teradial- calls one of those other files and you have not installed the entire package. So my first step here would be to go back and re-install the entire package:

net install st0444.pkg, replace

and try again.

If that doesn't resolve your problem, run

set tracedepth 1
set trace on
// YOUR teradial COMMAND HERE

That way, you will at least be able to see what command Stata couldn't recognize, and then you can pursue that issue.

[Gizem KAYA]

Mr. Schechter

Thank you for your help.
But when I run the code below, The STATA shows the following :

. net install st0444.pkg, replace
checking st0444 consistency and verifying not already installed...
all files already exist and are up to date.

. teradial y= x1 x2 x3
> x4, rts(crs) base(i) tename(te)

Radial (Debreu-Farrell) input-based measures of technical efficiency under
assumption of CRS technology are computed for the following data:

Number of data points (K) = 69
Number of outputs (M) = 1
Number of inputs (N) = 4

Data for reference set are not provided. Reference set is formed by 69 data
points, for which measures of technical efficiency are computed
unrecognized command
r(199);
Then I run the second one which you suggested, I get (I am copying the last part of the error, cause it is long):


> ovided {text} reference data points{p_end}"
}
- }
- if nref <= nO + nI {
capture noi error 409
display
if "`reference'" == "" & `refThere' == 0 {
display "{p 1 1 2 `ls'} {error}Number of {error}reference data points (" as result nr
> ef "{error}) must be larger than sum of number of outputs (" as result nO "{error}) a
> nd inputs(" as result nI "{error}){p_end}"
}
else {
display "{p 1 1 2 `ls'} {error}Number of {input} provided {error}reference data point
> s (" as result nref "{error}) must be larger than sum of number of outputs (" as resu
> lt nO "{error}) and inputs(" as result nI "{error}){p_end}"
}
matrix drop yobs xobs yxref yref xref
scalar drop nO nI nORef nIRef nref nobs rt ba
exit 409
}
- sca ifqh = 1
- matrix te = J(nobs, 1, .)
- if "`tename'" != ""{
= if "te" != ""{
- scalar intovar = 1
- plugin call te_radial `tename' if `touse', yobs xobs nO nI nobs yref xref nref rt ba
> ifqh intovar te
= plugin call te_radial te if __000001, yobs xobs nO nI nobs yref xref nref rt ba ifqh
> intovar te
unrecognized command
}
---------------------------------------------------------------------- end teradial ---
r(199);



When I check, there is no text like that " te if __000001" in the ado file.
And the "plugin call" code seems running.

Do you have any other suggestions ?

Thank you
Gizem

[Clyde Schechter]

So here's how to read this output. Look at the part

- plugin call te_radial `tename' if `touse', yobs xobs nO nI nobs yref xref nref rt ba 
> ifqh intovar te
= plugin call te_radial te if __000001, yobs xobs nO nI nobs yref xref nref rt ba ifqh 
> intovar te

Notice that it's two versions of almost the same thing. The first one is preceded by a single hyphen (-) character. That means it's the command as written in the .ado file (or wherever the code is coming from). Notice that it says `touse' after the -if-. The second version is preced by an = character. That is the version in which all of the local macros (in this case, tename and touse) have been replaced by what they contain. Comparing the two versions, we see that at this point Stata finds that tename contains te and touse contains __000001. So there is nothing remarkable about any of that.

With these macro substitutions made, Stata tries to execute the command and in doing so it determines that there is no command called -plugin-. But that's not true, unless you are running a very old version of Stata that predates plugins. So I can see two possibilities here.

It may be that the particular plugin being called, namely te_radial, is missing from your installation. I would try once again re-installing the package. This time do it more aggressively:

net install st0444.pkg, force

and see if that solves the problem.

If it does not, then there may be a problem with your Stata installation. I would make sure that you have the most current updates and executable by running -update all, force-.

If those things do not solve the problem, then I don't know what else to tell you, and I would recommend that you refer it to Stata Technical Support.

[Gizem KAYA]

Dear Sir,

I am very grateful to you four help. But, it did not work again.
I will try all these on another computer which has a new Stata version. If it does not work, I will refer it to technical support.