My Stata 16 does not show all the local variables that are supposed to appear in a table using esttab

[Francisco Olivero]

There is something in how Stata 16 is set in my computer that makes the following code not to work as it should:
eststo Agreementx_gets_face: reg Agreementx_gets Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) qui estadd local Sample="Face-to-Face" eststo Agreementx_gets_ele: reg Agreementx_gets Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) qui estadd local Sample="Electronic" eststo auto_exploitation_face: reg auto_exploitation Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) qui estadd local Sample="Face-to-Face" eststo auto_exploitation_ele: reg auto_exploitation Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) qui estadd local Sample="Electronic" esttab Agreementx_gets_face Agreementx_gets_ele auto_exploitation_face auto_exploitation_ele /// using "$latex/reg_x_earnings.tex", replace /// keep (Forced) /// b(2) se(2) star(* 0.10 ** 0.05 *** .01) /// nomtitles collabels(none) /// noobs nocons booktabs nonotes /// stats(N N_clust meanc Sample, fmt(%9.0fc 0 2 0) labels("Observations" "Num. Clusters" "Dep. Var. Mean" "Sample")) eststo clear
In the computer of my research assistant, the code works properly and I get this table:
{ \def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi} \begin{tabular}{l*{4}{c}} \toprule &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\ multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\ &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\ \midrule Forced & -115.01\sym{**} & 12.93 & 0.29\sym{***}& 0.01 \\ & (44.46) & (43.32) & (0.09) & (0.10) \\ \midrule Observations& 276 & 260 & 276 & 260 \\ Num. Clusters& 85 & 88 & 85 & 88 \\ Dep. Var. Mean& 1165.18 & 1172.06 & 0.42 & 0.42 \\ Sample &Face-to-Face & Electronic &Face-to-Face & Electronic \\ \bottomrule \end{tabular} }
However, when we run the code in my computer, I only get the local variable that I define first (whichever it is). The way the code is written, I get this (the last row is missing):
{ \def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi} \begin{tabular}{l*{4}{c}} \toprule &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\ multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\ &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\ \midrule Forced & -115.01\sym{**} & 12.93 & 0.29\sym{***}& 0.01 \\ & (44.46) & (43.32) & (0.09) & (0.10) \\ \midrule Observations& 276 & 260 & 276 & 260 \\ Num. Clusters& 85 & 88 & 85 & 88 \\ Dep. Var. Mean& 1165.18 & 1172.06 & 0.42 & 0.42 \\ Sample & & & & \\ \bottomrule \end{tabular} }
On the contrary, if I define first the Sample then the mean:
eststo Agreementx_gets_face: reg Agreementx_gets Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui estadd local Sample="Face-to-Face" qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) eststo Agreementx_gets_ele: reg Agreementx_gets Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui estadd local Sample="Electronic" qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) eststo auto_exploitation_face: reg auto_exploitation Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui estadd local Sample="Face-to-Face" qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) eststo auto_exploitation_ele: reg auto_exploitation Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1), vce(cluster Participant_id) qui estadd local Sample="Electronic" qui sum Agreementx_gets if e(sample)==1 qui estadd scalar meanc r(mean) esttab Agreementx_gets_face Agreementx_gets_ele auto_exploitation_face auto_exploitation_ele /// using "$latex/reg_x_earnings.tex", replace /// keep (Forced) /// b(2) se(2) star(* 0.10 ** 0.05 *** .01) /// nomtitles collabels(none) /// noobs nocons booktabs nonotes /// stats(N N_clust meanc Sample, fmt(%9.0fc 0 2 0) labels("Observations" "Num. Clusters" "Dep. Var. Mean" "Sample")) eststo clear
I get the opposite problem (the second to last row is missing):
{ \def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi} \begin{tabular}{l*{4}{c}} \toprule &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\ multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\ &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\ \midrule Forced & -115.01\sym{**} & 12.93 & 0.29\sym{***}& 0.01 \\ & (44.46) & (43.32) & (0.09) & (0.10) \\ \midrule Observations& 276 & 260 & 276 & 260 \\ Num. Clusters& 85 & 88 & 85 & 88 \\ Dep. Var. Mean& & & & \\ Sample &Face-to-Face & Electronic &Face-to-Face & Electronic \\ \bottomrule \end{tabular} }
This extends to other similar tables, and persists no matter whether I define variables as scalars, locals, or globals. Most importantly, it does not happen on the computer of my research assistants, so I guess it has to do with Stata settings in my computer.

I would very much appreciate your help.

[Nick Cox]

This was cross-posted at https://stackoverflow.com/questions/...d-to-appear-in

The version there (after my edits) is likely to be more readable than #1. Nevertheless, Francisco Olivero, please post a more readable version using CODE delimiters.

I suggested posting here. Nevertheless you are asked on Statalist to explain cross-posting.

As already stated on SO, I don't understand the question myself and I fail to see how we can guess what differs on two computers. I am more optimistic that experienced users of this suite may have an explanation in its terms.

[Andrew Musau]

The tables in SO are more clear. Here are two things that you can do:

1. Update your estout installation.

ssc install estout, replace

2. Associate the added locals or scalars to specific stored estimates. Here are illustrations for two sets of estimates.

eststo Agreementx_gets_face: reg Agreementx_gets Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1), vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc= r(mean): Agreementx_gets_face
qui estadd local Sample="Face-to-Face": Agreementx_gets_face

eststo Agreementx_gets_ele: reg Agreementx_gets Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1), vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc = r(mean): Agreementx_gets_ele
qui estadd local Sample="Electronic": Agreementx_gets_ele

[Andres Gago]

Eureka! Thanks for your answer Andrew Musau , I updated muy estout installation and now it works properly.

[Francisco Olivero]

Here it's the correction with code delimiters:

There is something in how Stata 16 is set in my computer that makes the following code not to work as it should:

eststo Agreementx_gets_face: reg Agreementx_gets Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)
qui estadd local Sample="Face-to-Face"

eststo Agreementx_gets_ele: reg Agreementx_gets Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)
qui estadd local Sample="Electronic"

eststo auto_exploitation_face: reg auto_exploitation Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)
qui estadd local Sample="Face-to-Face"

eststo auto_exploitation_ele: reg auto_exploitation Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)
qui estadd local Sample="Electronic"

esttab Agreementx_gets_face Agreementx_gets_ele auto_exploitation_face auto_exploitation_ele ///
using "$latex/reg_x_earnings.tex", replace /// keep (Forced) /// b(2) se(2) star(* 0.10 ** 0.05 *** .01) ///
nomtitles collabels(none) /// noobs nocons booktabs nonotes
/// stats(N N_clust meanc Sample,
fmt(%9.0fc 0 2 0) labels("Observations" "Num. Clusters" "Dep. Var. Mean" "Sample"))

eststo clear

In the computer of my research assistant, the code works properly and I get this table:

{ \def\sym#1{\ifmmode^{#1}\else^{#1}^{#1}\fi}
\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\            
&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}
&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule
Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&     1165.18         &     1172.06         &        0.42         &        0.42         \\
Sample      &Face-to-Face         &  Electronic         &Face-to-Face         &  Electronic         \\
\bottomrule
\end{tabular}

\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\
            &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule
Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&     1165.18         &     1172.06         &        0.42         &        0.42         \\
Sample      &Face-to-Face         &  Electronic         &Face-to-Face         &  Electronic         \\
\bottomrule \end{tabular}
}

However, when we run the code in my computer, I only get the local variable that I define first (whichever it is). The way the code is written, I get this (the last row is missing):

{ \def\sym#1{\ifmmode^{#1}\else^{#1}^{#1}\fi}
\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\
            &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&     1165.18         &     1172.06         &        0.42         &        0.42         \\
Sample      &                     &                     &                     &                     \\
\bottomrule
\end{tabular}

\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\
            &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule
Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&     1165.18         &     1172.06         &        0.42         &        0.42         \\
Sample      &                     &                     &                     &                     \\
\bottomrule \end{tabular}
}

On the contrary, if I define first the Sample then the mean:

eststo Agreementx_gets_face: reg Agreementx_gets Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui estadd local Sample="Face-to-Face"
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)

eststo Agreementx_gets_ele: reg Agreementx_gets Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui estadd local Sample="Electronic"
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)

eststo auto_exploitation_face: reg auto_exploitation Forced i.round_dummy if face==1 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui estadd local Sample="Face-to-Face"
qui sum Agreementx_gets if e(sample)==1
qui estadd scalar meanc r(mean)

eststo auto_exploitation_ele: reg auto_exploitation Forced i.round_dummy if face==0 & (Forced==1 | Baseline==1),
vce(cluster Participant_id)
qui estadd local Sample="Electronic"
qui sum Agreementx_gets if e(sample)==1
qui estadd
scalar meanc r(mean)

esttab Agreementx_gets_face Agreementx_gets_ele auto_exploitation_face auto_exploitation_ele
/// using "$latex/reg_x_earnings.tex", replace /// keep (Forced) /// b(2) se(2) star(* 0.10 ** 0.05 *** .01)
/// nomtitles collabels(none) /// noobs nocons booktabs nonotes
/// stats(N N_clust meanc Sample,
fmt(%9.0fc 0 2 0) labels("Observations" "Num. Clusters" "Dep. Var. Mean" "Sample"))
eststo clear

I get the opposite problem (the second to last row is missing):

{ \def\sym#1{\ifmmode^{#1}\else^{#1}^{#1}\fi}
\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\
            &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule
Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&                     &                     &                     &                     \\
Sample      &Face-to-Face         &  Electronic         &Face-to-Face         &  Electronic         \\
\bottomrule
\end{tabular}

\begin{tabular}{l*{4}{c}} \toprule             &\multicolumn{1}{c}{(1)}&\multicolumn{1}{c}{(2)}&\  multicolumn{1}{c}{(3)}&\multicolumn{1}{c}{(4)}\\
            &\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{Decider's Profits}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}&\multicolumn{1}{c}{$\mathds{1}($ Decider's Profits$<x)$}\\
\midrule
Forced      &     -115.01\sym{**} &       12.93         &        0.29\sym{***}&        0.01         \\
            &     (44.46)         &     (43.32)         &      (0.09)         &      (0.10)         \\
\midrule
Observations&         276         &         260         &         276         &         260         \\
Num. Clusters&          85         &          88         &          85         &          88         \\
Dep. Var. Mean&                     &                     &                     &                     \\
Sample      &Face-to-Face         &  Electronic         &Face-to-Face         &  Electronic         \\
\bottomrule
\end{tabular}
}

This extends to other similar tables, and persists no matter whether I define variables as scalars, locals, or globals. Most importantly, it does not happen on the computer of my research assistants, so I guess it has to do with Stata settings in my computer.

I would very much appreciate your help.

[Nick Cox]

This thread is confusing. From comments on Stack Overflow (which may or may not disappear soon) it seems that

1. Francisco Olivero and Andres Gago are co-workers.

2. The solution in #3 from Andrew Musau solved the problem (#4).

3. So #5 is just a repost of #1 and there is nothing else to discuss.