sideways option in nested regressions

In nested regression models, asdocx reports statistics showing statistical significance in the second row, below the regression coefficients. In the recent update (version 2.2.1: June 20, 2022) , I have added the sideways option to report significance statistics (such as t-values, p-values, standard errors) or confidence intervals sideways to the right of the regression coefficients.

In this post, I’ll show several examples using the sideways option. The first example tries to make a table is is often required in medical / epidemiological studies.

Example 1: Regression table to report confidence intervals to the right-side

In medical / epidemiological studies, researcher often report odd ratios and confidence intervals in a single row. To make such a table, use the following options of asdocx:

nest : invoke a nested regression table

rep(ci) :  to report confidence

sideways : to report the confidence intervals to the right side of the odd ratios

eform : to report odd rations (exponentiated coefficients)

dec(3) : set three decimal places

tzok: report equal number of decimal points, even if the trailing values are all zeros.

 webuse lbw , clear
 
 asdocx logit low age lwt i.race smoke ptl ht ui , ///
        nest sideways tzok rep(ci)

 asdocx logit low lwt i.race smoke ptl ht ui , ///
        nest sideways tzok rep(ci)