One major drawback in a fixed effects ordered logit model (FE OL) is its inability to compute marginal effects due to unknown cut point differences. Previous literatures collapsed the dependent variable to a set of binary variables, then apply Chamberlain's (1980) conditional MLE to each of these binary choice model, and finally combine the information.
A new estimator by Muris (2016) was introduced to estimate the differences in the cut points along with the regression coefficient, hence providing bounds for a marginal effect (see Muris (2016) section 3.3).
The attached link at the bottom is the codes to replicate the tables in Muris's (2016) paper. After the transformation of the dependent variable, the new estimator can be obtained using the clogit command.
Obviously clogit does not allow the computation of margins.
With very basic knowledge in Stata programming, I need suggestion on how to compute margins in this case, using the information of the cut point differences to bound the marginal effects. If I want to write my own program (to compute the margins), where do I start? Thank you.
References:
Muris, C. (2016). Estimation in the fixed effects ordered logit model. Review of Economics and Statistics.
Codes.
Comments on how to improve this question is highly appreciated.
A new estimator by Muris (2016) was introduced to estimate the differences in the cut points along with the regression coefficient, hence providing bounds for a marginal effect (see Muris (2016) section 3.3).
The attached link at the bottom is the codes to replicate the tables in Muris's (2016) paper. After the transformation of the dependent variable, the new estimator can be obtained using the clogit command.
Code:
clogit y-transformed xvar i.cut, group(newid) cluster(caseid) lincom 3.cut-2.cut lincom 4.cut-3.cut
With very basic knowledge in Stata programming, I need suggestion on how to compute margins in this case, using the information of the cut point differences to bound the marginal effects. If I want to write my own program (to compute the margins), where do I start? Thank you.
References:
Muris, C. (2016). Estimation in the fixed effects ordered logit model. Review of Economics and Statistics.
Codes.
HTML Code:
https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/IUJ3VZ
Comment