Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Adding significance stars to linear combination test results

    Hi,


    I have saved the results of two linear combination tests as scalars (the coefficients and standard errors), and I am using estab to export my results into Latex using the following code:

    Code:
    
    forvalues num = 1/6 {
        xi: avg_effect `var`num'', x(NVH TIt3 CMt3 i.cluster_code_new) effect (NVH TIt3 CMt3) control (NVH==0) r cluster(cluster_code_new) keepmissing
        estimates store spec5ia_`num'
        estadd local fixed "Yes" , replace
        lincom ae_TIt3+ae_NVH
        estadd scalar ln1mean r(estimate)
        estadd scalar ln1se r(se)
        est restore spec5ia_`num'
        lincom ae_CMt3+ae_NVH
        estadd scalar ln2mean r(estimate)
        estadd scalar ln2se r(se)
    }
    
    # delimit ;
    esttab      spec5ia_*
    using "spec5i_applicants.tex", 
            keep(ae_NVH
                ae_TIt3
                ae_CMt3)
            order(ae_NVH
                ae_TIt3
                ae_CMt3)
            cells(b(fmt(%5.3f) star) se(fmt(%5.3f) par))
            t(par fmt(2))
            starlevels(* .10 ** .05 *** .01)     
            mtitles("\shortstack{\textbf{HH Work}\\\textbf{}}"
                    "\shortstack{\textbf{Education}\\\textbf{}}"
                    "\shortstack{\textbf{HH Welfare}\\\textbf{}}"
                    "\shortstack{\textbf{Confidence}\\\textbf{}}"
                    "\shortstack{\textbf{Political}\\\textbf{Participation}}"
                    "\shortstack{\textbf{Empowerment}\\\textbf{}}")
             varlabels(ae_NVH "Village Treatment Applicants (VTA)"
             ae_TIt3 "VTA*Trainee Info"
             ae_CMt3 "VTA*Community mobilization")        
            refcat(ae_NVH " \multicolumn{7}{l}{\textbf{Panel A: Applicants}} \\ \midrule ", nolabel)
            stats(ln1mean ln1se ln2mean ln2se N fixed, labels(
                  "\hline VTA+VTA*Trainee Info" " " "VTA+VTA*Community Mobilization" " "  "\hline Observations" "Village Fixed Effects?")     
            fmt(3 3 3 3 0)
            star(ln1mean)
            layout(@ (@) @ (@) @ @))
            nonumbers
            booktabs
            fragment
            noobs
            nonotes
            collabels(none)
            replace;

    How can I get the significance starts to show for the results of the lncom tests in my table? The approach suggested in this post did not work for me.

    Thanks.

Working...
X