Test hypothesis in transition matrix

[Isidora Vergara]

Hi Statalist,

I am doing a research about employment transitions and I have a data of this type (this is a simple example):

clear
input float(id date_q year month quarter sex status) byte l_status float(strata psu)
1 232 2018  2 1 1 1 . 1 1
1 233 2018  5 2 1 2 1 1 1
2 233 2018  5 2 1 2 . 1 1
2 234 2018  8 3 1 2 2 1 1
3 236 2019  2 1 2 1 . 1 1
3 237 2019  5 2 2 2 1 1 1
3 238 2019  8 3 2 2 2 1 1
4 236 2019  2 1 1 2 . 1 1
5 242 2020  8 3 2 1 . 1 1
5 243 2020 11 4 2 1 1 1 1
6 242 2020  8 3 1 1 . 1 1
6 243 2020 11 4 1 2 1 1 1
7 241 2020  5 2 2 2 . 1 1
7 242 2020  8 3 2 2 2 1 1
7 243 2020 11 4 2 2 2 1 1
end
format %tq date_q
label values sex sex
label def sex 1 "Men", modify
label def sex 2 "Women", modify
label values status status
label def status 1 "Unemployed", modify
label def status 2 "Employed", modify
label values l_status l_status
label def l_status 1 "Unemployed", modify
label def l_status 2 "Employed", modify

After defining the sample characteristics (with the svyset command), I estimated the transition probabilities with these commands:

estpost svy, subpop(if sex==1): tabulate l_status status, row se format(%11.3f) // For men.
estpost svy, subpop(if sex==2): tabulate l_status status, row se format(%11.3f) // For women.

But my doubt is that I would like to be able to compare the obtained probabilities for men and women (e.d., to be able to say if one transition is more or less likely for man than for women) but I do not know how to do this. Could you help me please?

I am using Stata 17.

Thanks in advance,
Isidora.