Conducting Interaction Analysis with Two Continous Variables in the Context of Ologit Analysis

[Ferdi Djajadisastra]

Can anyone point me to a user-friendly guide to generate the interaction figure in the context of ordered logistic analysis? I am conducting an interaction analysis with two continuous variables and a 4-level categorical variable as the outcome. I appreciate everyone's help!

[Carlo Lazzaro]

Ferdi:

. use https://www.stata-press.com/data/r19/fullauto.dta
(Automobile models)

. ologit rep78 c.mpg##c.weight

Iteration 0:  Log likelihood = -93.692061  
Iteration 1:  Log likelihood = -83.875222  
Iteration 2:  Log likelihood = -83.305793  
Iteration 3:  Log likelihood = -83.298258  
Iteration 4:  Log likelihood = -83.298256  

Ordered logistic regression                             Number of obs =     69
                                                        LR chi2(3)    =  20.79
                                                        Prob > chi2   = 0.0001
Log likelihood = -83.298256                             Pseudo R2     = 0.1109

--------------------------------------------------------------------------------
         rep78 | Coefficient  Std. err.      z    P>|z|     [95% conf. interval]
---------------+----------------------------------------------------------------
           mpg |   .4618658   .1613006     2.86   0.004     .1457225    .7780091
        weight |    .002155   .0011177     1.93   0.054    -.0000357    .0043456
               |
c.mpg#c.weight |  -.0001418   .0000531    -2.67   0.008    -.0002459   -.0000377
---------------+----------------------------------------------------------------
         /cut1 |   3.864992   3.965762                     -3.907759    11.63774
         /cut2 |   5.637504   3.921177                     -2.047863    13.32287
         /cut3 |   8.033101   3.938183                      .3144052     15.7518
         /cut4 |   9.806465   4.019011                      1.929348    17.68358
--------------------------------------------------------------------------------

.

See -fvvarlist- for further details.

[Felix Bittmann]

I am not exactly sure what you want to produce with your analysis. This tutorial here offers some nice examples: https://stats.oarc.ucla.edu/stata/fa...ion-in-ologit/ Wile one of the interaction variables is categorical and not continuous, it explains some central concepts.