How to bootstrap Malmquist Producitivty Index?

[Fanny Goude]

Dear Stata users,

I'm using the malmq command to produce the Malmquist Productivity Index. However, I wish to bootstrap the index and its' components and calculate standard errors and confidence intervals. How can I do this? I'm using Stata 15.1.

Thanks,
Fanny Goude

[Serena Chang]

Hi,

I just saw an article 'Analysing Productivity Changes Using the Bootstrapped Malmquist Approach: The Case of the Iranian Banking Industry'. I wish it may give you some hints.

[Cord-Friedrich von Hobe]

Dear Stata Users,

I have the same problem as Fanny. I have 86 DMUs with 6 input variables and one output variable over 5 years. I want to bootstrap Malmquist indices and its components and calculate the confidence intervals for them afterwards. I've tried writing a Loop for it:

---------------------------------------------------------------------------------
capture program drop malmq_wrap

program define malmq_wrap, rclass

malmq input1 input2 input3 input4 input5 input6 = output, ort(i) period(year)

tempname W

matrix `W' = r(prodidxrslt)

forvalues i = 1/344 {

return scalar result`i' = `W'[`i', 2]

}

exit

end



local results

forvalues i = 1/344 {

local results `results' result`i' = r(result`i')

}



bootstrap `results', rep(1000): malmq_wrap

---------------------------------------------------------------------------------------

I always receive the following error: "insufficient observations to compute bootstrap standard errors no results will be saved"

Does anybody know what I'm doing wrong or how to calculate the bootstrap scores and the confidence intervals correctly?

Thank you very much!

Cord