stargazer | Get Slack notifications whenever someone stars your Github | Chat library

You can store the output in a list -

• the stray lines below and above the table are caused by all these \\[-1.8ex] used even though there is no line to finish

• the stray lines on the right comes from a mismatch of how many columns there are actually in the table (8) and how many columns you tell latex there would be in the table (9).

• please reconsider such a table layout. Using vertical lines does not exactly look like a professional done table, see the booktabs package documentation for further inspiration http://mirrors.ctan.org/macros/latex/contrib/booktabs/booktabs.pdf

• please also have a look at the siunitx package to correctly align and format your numbers. At the very minimum, you shouldn't abuse hyphens as minus signs.

• If your table is already too wide to fit in the available text area, the last thing you should do is adding extra wide space between the columns

Your fixed MWE:

I asked for clarification in a comment, but here is my best attempt at guessing what you want in the table:

1. Multiple models side-by-side
2. Model estimates
3. p values in parentheses below estimates
4. Kernel type at the bottom of hte table
5. Bandwidth selection at the bottom of the table

To achieve this, I only modified your glance.rdrobust method, and I used the statistic argument of the modelsummary function.

Load libraries and define custom tidy and glance methods to extract information (see documentation) from rdrobust objects:

The problem is in the line

arquivo_json = json.dumps(registro_json, indent=2, sort_keys=False)

Which according to the documentation, json.dumps "Serializes obj to a JSON formatted str according to conversion table"

In effect, the problem is that you are serializing the registro_json object twice, and ending up with a str. If you remove the offending line and directly pass registro_json to the gravar_arquivo_json function, everything should work.

Updated code:

Answer

The package does string matching to abond_2step$call[1]. In your case, that would be plm::pgmm(). The package is hardcoded to look for pgmm(), and thus doesn't properly set it.

Another example of code that doesn't work

There is a loose cat() call, followed by another cat() inside if/else which doubles the output. Comment that line out:

Final answer:

You can use the modelsummary package for R to save a correlation table directly to a Word document. (Disclaimer: I am the author.) Using the datasummary_correlation function and its output argument, we have:

I do not know how to achieve this in stargazer, but it is very easy to do with the modelsummary package for R (disclaimer: I am the author).

modelsummary supports 100+ models out-of-the-box, but not model objects of class pool_mi, which is what your code produces. Fortunately, it is very easy to add support for new models, as described in detail in the documentation.

Specifically, we need to define two S3 methods called tidy.CLASSNAME and glance.CLASSNAME. The first method returns a data.frame with one coefficient per row and with column names that follow the standard terminology from the broom package. The second method returns a one-row data.frame with goodness-of-fit or other model characteristics, one per column.

In your case, these simple methods seem to do the job (obviously, you can customize):