It's often necessary to produce an output table that doesn't fit any of the format provided by the standard tools outreg, esttab, etc.. Some examples:

• Showing p-values for linear combinations of coefficients
• Putting certain coefficients in bold
• Putting different outcome variables in the same row of a regression table
• Multi-panel tables with different formatting in each panel

Stata-tex allows you to separate the LaTeX table template from the table data. This lets you set up and compile exactly the LaTeX table you want, with placeholders for the data. Then you can generate the data separately, and transfer it into the LaTeX table automatically.

It takes a bit of time to get right on the first time, but everyone on our team now loves this tool and relies on it for tables of any complexity. We find it far easier to just write the template we want rather than trying to figure out esoteric estout or esttab parameters.

• Arbitrary customization of Stata/LaTeX tables
• Iterate on the LaTeX table without regenerating estimates
• Stata code is much much cleaner
• Can get LaTeX table looking right without using Stata
• Easy to copy table templates to new contexts
• One regression can create estimates that are used in many tables

Create a latex table, with placeholders for all the numbers you want to put in later. Placeholders are marked with $$, e.g. $$beta1$$. So you create a regular latex table, with lines like this:

Treatment           &     $$treatment1_starbeta$$  &     \textbf{$$treatment2_starbeta$$}  &     $$treatment3_starbeta$$  &     \textbf{$$treatment4_starbeta$$}  \\
                    &     $$treatment1_se$$        &     \textbf{$$treatment2_se$$}        &     $$treatment3_se$$        &     \textbf{$$treatment4_se$$}        \\

In this example, treatment1 and treatment2 are estimates from different regressions -- treatment1_beta is a placeholder for the first regression estimate, and treatment1_se for the standard error. Notice that estimates in the second and forth column are customized to be in bold face.

Finish this table (i.e. with tabular etc.), until it compiles in LaTeX in the format you want. The package includes a very simple sample table in output_table.tex.

In Stata, use the post-estimation command store_est_tpl (included in the stata-tex package), to generate a CSV file with all the estimation results:

reg mpg t [...]
store_est_tpl using sample_table.csv, coef(t) name(treatment2) all

This will append a block to sample_table.csv containing the following lines (with example numbers):

treatment2_p, 0.00
treatment2_beta, 0.009
treatment2_starbeta, 0.009***
treatment2_se, 0.001
treatment2_n, 147143
treatment2_r2, 0.00

Note that the placeholder beta holds just the beta coefficient, while starbeta calculates a p-value and shows stars for p<0.1, p<0.05, p<0.01. e.g. beta contains 0.06, and starbeta contains 0.06**.

store_est_tpl takes an optional format() parameter that lets you specify a different format for the coefficient and standard error (e.g. "%5.2f"). The default is 3 decimal points. p-values and r2 always have 2 decimal points.

The all parameter is short for beta se n p r2 -- the latter let you choose exactly what estimation statistics you want to output. Outputting statistics you don't use is fine, so all is usually the right answer.

You can run store_est_tpl multiple times following a single estimation if you want to store multiple coefficients. If you want to store some other value in this file, e.g. a p-value from an F test or significance test for a difference between two coefficients, you can store an arbitrary string using:

insert_into_file using sample_table.csv, key(treatment2_ftest) value(0.35) format(%5.2f)

Note that both store_est_tpl and insert_into_file will replace any line in the CSV file that currently has the same coefficient name.

Finally, use table_from_tpl to transfer the estimates into the LaTeX template (via Python):

table_from_tpl, t(treatment_tpl.tex) r(sample_table.csv) o(output_table.tex) 

t = template file, r = replacement data file, o = output file

You can also optionally add or suppress significance stars from the output file with the following two commands:

table_from_tpl, t(treatment_tpl.tex) r(sample_table.csv) o(output_table.tex) add_stars
table_from_tpl, t(treatment_tpl.tex) r(sample_table.csv) o(output_table.tex) drop_stars

This operates by replacing beta with starbeta in the template (or vice versa) when generating the output file. If you have stars on some coefficent name other than beta, you will need to create custom templates with and without stars.

Place all stata-tex files in the same folder, including the sample template treatment_tpl.tex. test.do provides a complete example using the standard auto dataset. It creates two output files: output_table.tex and output_table_no_stars.tex.

• Unlike outreg, estout, etc., you can't add and remove columns from these tables without modifying the LaTeX template. This is an inherent limitation of having a totally customizable table template.

• If a placeholder appears multiple times in the output file, only the first appearance will be used. However, if you are only using insert_into_file and store_est_tpl, you should not run into this as these programs replace existing lines with the same parameter values (as of 10/30/2018).

• The templates don't compile themselves, because latex interprets the replacement markers $$ and _ as special characters. I couldn't find any clean way around this, but let me know if you have a better idea for how to tag replacement strings.

table_from_tpl() needs to find table_from_tpl.py either in the current folder, or in the folder specified by the Stata global macro $PYTHONPATH.

This has been tested most thoroughly with Python 2.7. If you encounter difficulties with other versions of Python, please let me know.