Ologit, odd ratios export

[Enrico Azzini]

Hi, I'm performing an ordered logist regression:

 eststo: quietly ologit In_sat i.migrant_groupw1 fulltime employee c.age##c.age i.education i.sex i.activity i.occupation i.Married i.year i.Countryoforign i.region_name if working==1, vce(cluster region_name) or

However when I use esttab to export the results the coefficient reported ar not in odd ratios.
I tryed to perform the regression without eststo

 ologit In_sat i.migrant_groupw1 fulltime employee c.age##c.age i.education i.sex i.activity i.occupation i.Married i.year i.Countryoforign i.region_name if working==1, vce(cluster region_name) or

and I optain odd ratios, Can you please me help to export the output in odd ratios?
Thank you for your attention

[Andrew Musau]

estout is from SSC, as you are asked to explain (FAQ Advice #12).


\(\text{Odds Ratio}\; = \; \text{exp(log odds ratio)}\), so you need to add the option

eform

to the esttab command.

[Oskar Lindgren]

Jumping in on this thread. Andrew Musau if one doesn't add the eform option, what results are reported when using the esttab command?

[Andrew Musau]

The ordered logit coefficients. These represent the change in the log-odds of being in a higher category of the dependent variable associated with a one-unit change in the predictor variable, holding all other variables constant.

[Oskar Lindgren]

Continuing here, perhaps you have a simple solution to this too Andrew Musau. I have survey data where my dependent variable is ordinal with five levels (policy_sup) and a number of independent variables that I would like to assess the effects of. I'm running a ordered logistic model which works fine and I get the odds ratios through or. However, I would like to display the estimates of the independent variables on the dependent through average marginal effects. I first runt the ologit command.

 ologit policy_sup x1 x2 x3 x4 x5 [pweight=weight], robust

This generates one logit coefficient for each independent variable.

--------------------------------------------------------------------------------
| Robust
fuel_rationing | Coefficient std. err. z P>|z| [95% conf. interval]
---------------+----------------------------------------------------------------
x1 |
x2 |
x3 |
x4 |
x5 |

---------------+----------------------------------------------------------------
/cut1 |


However, when I want to turn these logit coefficients into average marginal effects:

 margins, dydx(*)

Stata provides predictions for each category of the depvar, while I would like one estimate for each independent variable, just as if I would run the ologit command with the or option. Sorry if this is completely obvious (or impossible), but how do I get average marginal effects without Stata providing these for each level of the depvar?

Best,