xtoaxaca decomposition and dfl

[Vivian Phan]

Dear ALL

In practice, I have data for 3 years 2000, 2010, 2020. I would like to get the detailed decomposition of the gender wage gap across 3 years. I also want to put the male wage distribution of the middle year (i.e. 2010) as the counterfactual wage distribution. While doing the research, I find the xtoaxaca and dfl command in stata very useful to address my question. However, it is my first time to know the xtoaxaca, so when I run the code xtoaxaca, I find the code is wrong

 reg logwage age education female i.year, r
est sto m1
xtoaxaca age education, groupvar(female) groupcate (0 1) timevar(year) times(2000 2010 2020) model(m1) timeref(2010) change(mpjd) noisily forcesample

[ Progress ] Calculating mean outcomes ...
[ Progress ] Calculating coefficients ...

[ Progress ] Running level decomposition ...
(xtoaxaca_mata_levels() in lxtoaxaca, compiled by Stata 16.1, is too new to be run by this
version of Stata and so was ignored)
                 <istmt>:  3499  xtoaxaca_mata_levels() not found
r(3499);

I wonder how I can get it right.

Looking forward to get some help.

Many thanks

[Jared Greathouse]

Buy Stata 16.

[Vivian Phan]

Hi Jared

I think I tried this with Stata 16 and still have got the same problem, Any ideas?

Many thanks

[FernandoRios]

Hi Vivian,
Just as Jared suggests you need Stata v16.1. In other words, you probably need to either update your current version of Stata (update all).
There are other options:
1. You could try to dig into the xtoaxaca ado files (all of them), and comment out all instances where they put "version XX.X". This may allow the command to work in your computer, but may also create other things to break.
2. You could try contacting the authors, THey may provide help or have a version of the command that works under older Stata versions.
HTH