CODEWITHSUNDEEP
rlatexstargazertexreg
User981636
User981636

Reputation: 3629

Exporting R output into Latex - Stargazer for non suported objects

I'm estimating models in R using the frontier package and I need to export the results into Latex. The output is quite similar to a lm regression [see below] but frontier objects are not supported by stargazer to export them into Latex code. Is there any way to work around this? Any idea? *I am also looking into texreg and apsrtable, so far unsuccessfully.

Example of frontier regression output:

enter image description here

Upvotes: 2

Views: 1605

Answers (1)

Benjamin
Benjamin

Reputation: 17279

I don't know much about getting stargazer to output unsupported models, but you can use atidy method from the broom package to get the basic output into a format compatible with xtable, knitr::kable, or pixiedust

library(broom)
library(frontier)

# example included in FRONTIER 4.1 (cross-section data)
data( front41Data )

# Cobb-Douglas production frontier
cobbDouglas <- sfa( log( output ) ~ log( capital ) + log( labour ),
                    data = front41Data )
tidy(cobbDouglas, conf.int = TRUE)
broom:::tidy.lm(cobbDouglas)

          term  estimate  std.error statistic      p.value
1  (Intercept) 0.5616193 0.20261685  2.771829 5.574228e-03
2 log(capital) 0.2811022 0.04764337  5.900132 3.632107e-09
3  log(labour) 0.5364798 0.04525156 11.855499 2.015196e-32
4      sigmaSq 0.2170003 0.06390907  3.395454 6.851493e-04
5        gamma 0.7972069 0.13642438  5.843581 5.109042e-09

For the summary statistics, you would need to write your own glance method, as the frontier objects aren't compatible with broom:::glance.lm.

But I think the end story is that, if you want to mimic the stargazer output, you'll have to do some preprocessing work.

And since I'm feeling ambitious today, here's a tidy method for frontier objects.

tidy.frontier <- function(x, conf.int = FALSE, conf.level = .95,
                          exponentiate = FALSE, quick = FALSE, ...)
{
  broom:::tidy.lm(x, conf.int = conf.int, conf.level = conf.level,
                  exponentiate = exponentiate, quick = quick, ...)
}


# example included in FRONTIER 4.1 (cross-section data)
data( front41Data )

# Cobb-Douglas production frontier
cobbDouglas <- sfa( log( output ) ~ log( capital ) + log( labour ),
                    data = front41Data )
tidy(cobbDouglas, conf.int = TRUE)

Upvotes: 2

Related Questions