CODEWITHSUNDEEP
r-markdownstargazertexreg
ModalBro
ModalBro

Reputation: 554

Outputting the results from bife object to Latex in Rmarkdown?

I'm estimating a fixed-effects probit model using the bife package in R. I'm trying to extract the output into something I can use with either stargazer or texreg so I can output them into a paper using Rmarkdown to create a LaTeX object. I'm aware I can manually extract the coefficients and standard errors, etc., but I'm wondering if there isn't a more efficient way to coerce this object into something that'd work with either package.

Here's a reproducible example: 

install.packages("bife")
library(bife)
data("iris")
iris$big <- ifelse(iris$Sepal.Length > median(iris$Sepal.Length),1,0)
output <- bife(big ~ Sepal.Width + Petal.Length | Species, data=iris, "logit")

Upvotes: 1

Views: 568

Answers (1)

Barbusse
Barbusse

Reputation: 11

I think I found an alternative solution for this one, even if it is probably too late

Basically, first, I went on the repository of the package "texreg" and found this function: 

extract.bife <- function(model,
                         include.loglik = TRUE,
                         include.deviance = TRUE,
                         include.nobs = TRUE,
                         ...) {
  s <- summary(model)
  coefficient.names <- rownames(s$cm)
  co <- s$cm[, 1]
  se <- s$cm[, 2]
  pval <- s$cm[, 4]

  gof <- numeric()
  gof.names <- character()
  gof.decimal <- logical()
  if (include.loglik == TRUE) {
    lik <- logLik(model)
    gof <- c(gof, lik)
    gof.names <- c(gof.names, "Log Likelihood")
    gof.decimal <- c(gof.decimal, TRUE)
  }
  if (include.deviance == TRUE) {
    gof <- c(gof, deviance(model))
    gof.names <- c(gof.names, "Deviance")
    gof.decimal <- c(gof.decimal, TRUE)
  }
  if (include.nobs == TRUE) {
    n <- s$nobs["nobs"]
    gof <- c(gof, n)
    gof.names <- c(gof.names, "Num. obs.")
    gof.decimal <- c(gof.decimal, FALSE)
  }

  tr <- createTexreg(
    coef.names = coefficient.names,
    coef = co,
    se = se,
    pvalues = pval,
    gof.names = gof.names,
    gof = gof,
    gof.decimal = gof.decimal
  )
  return(tr)
}

So for your example, just apply it on your model and use the function texreg and you may have a Latex-"like" output

tr <- extract.bife(output)
texreg(tr)

I hope it will help! Best

Upvotes: 1

Related Questions