How can I determine if these predicted probabilities are significantly different from each other in the table below with Stata?

[smith Jason]

Hi, I have a small dataset for the purpose of demonstration below, which is about a table of the disciplinary action probabilities that students from different ethnic groups may face after different levels of disciplinary infractions. What I need now is to run Stata code to test if these predicted probabilities are significantly different from each other.
I am not sure how to write the code, as this table is the result of running an ordinal longitudinal model and obtaining the results using the "margins" command.
race--1(black);2-hispanic;3-White;4-Other
outcome-1 (receiving minor disciplinary);2-(receiving medium disciplinary action);3- (receiving severe disciplinary action)
infraction1-(the predicted probability of receiving minor disciplinary action); infractioninfraction2- (the predicted probability of receiving medium disciplinary action); infraction3- (the predicted probability of receiving severe disciplinary action)

clear
input byte race outcome infraction1 infraction2 infraction3
1 1 0.674 0.698 0.228
2 1 0.690 0.465 0.125
3 1 0.712 0.577 0.290
4 1 0.718 0.547 0.147
1 2 0.284 0.388 0.528
2 2 0.255 0.390 0.367
3 2 0.240 0.329 0.435
4 2 0.235 0.353 0.390
1 3 0.069 0.146 0.353
2 3 0.056 0.148 0.510
3 3 0.050 0.094 0.276
4 3 0.048 0.111 0.417
end

Can someone help me with this problem? Thank you!

[smith Jason]

clear
input byte race outcome byte infraction_level double infraction
1 1 1 0.674
1 1 2 0.698
1 1 3 0.228
1 2 1 0.284
1 2 2 0.388
1 2 3 0.528
1 3 1 0.069
1 3 2 0.146
1 3 3 0.353
2 1 1 0.690
2 1 2 0.465
2 1 3 0.125
2 2 1 0.255
2 2 2 0.390
2 2 3 0.367
2 3 1 0.056
2 3 2 0.148
2 3 3 0.510
3 1 1 0.712
3 1 2 0.577
3 1 3 0.290
3 2 1 0.240
3 2 2 0.329
3 2 3 0.435
3 3 1 0.050
3 3 2 0.094
3 3 3 0.276
4 1 1 0.718
4 1 2 0.547
4 1 3 0.147
4 2 1 0.235
4 2 2 0.353
4 2 3 0.390
4 3 1 0.048
4 3 2 0.111
4 3 3 0.417
end

[smith Jason]

The 2nd floor is the long format of the data converted from wide format.

[smith Jason]

I feel frustrated and nobody could help me....................

[Maarten Buis]

That is a bit quick to feel frustrated. Maybe consider that people live in different time-zones...

Just use the post option with margins and after that you can run a regular test with test command.

[smith Jason]

Thank you. But I don't know how to do this. Could you please help me with the code?

[smith Jason]

Can someone help me?