Unconditional quantile regression after multiple imputation

[Mamadou DIALLO]

Hi,

Can you please help me with the command to run an unconditional quantile regression after multiple imputation ?
Thank you in advance.

[FernandoRios]

You can use rifhdreg
mi estimate, cmdok: rifhdreg y x z, rif(q(10))

[Mamadou DIALLO]

Fernando,
The code works well.
Thank you!

[Mamadou DIALLO]

Now I would like to do the Oaxaca-Blinder decomposition after the unconditional quantile regression. Can you help me to do this decomposition with stata.

[FernandoRios]

technically you should be able to use oaxaca_rif as well.
But I never tried that

[Mamadou DIALLO]

Hello,
I would like to do a median comparison test, but my code does not work. Here is the code : mi estimate, cmdok vartable : ranksum patrimoine_nette, by(sex) porder.
Can you help me?

[Rich Goldstein]

first, ranksum is not a test of medians and, second, it is not an estimation command it is a test; your project is not at all clear to me, but maybe you want qreg at the 50th percentile?