regtable: Format R regression output
In yixinsun1216/catwalk: Showing Off Models

Description Usage Arguments Examples

This routine creates Latex code, HTML code, and text tables that present regression output.

regtable(ms, est, mnames = NULL, est_names = NULL, extra_rows = NULL,
  se_fun = vcov, stats = c("r.squared", "adj.r.squared"),
  stats_names = c("$R^2$", "Proj. $R^2$"), n_obs = TRUE,
  output_format = "latex", header = NULL, sig_stars = FALSE,
  decimals = NULL, note = NULL, ...)

`ms`	A list of one or more model objects that are compatible with the summary() function.
`est`	A character vector of the covariates to output. To output different coefficients for different models, pass in est as a list, where est[[i]] specifies the covariates to output for ms[[i]].
`mnames`	A character vector of labels for each model object in ms
`est_names`	A character vector of labels for est. If outputting different covariates for different models, pass in est_names as a list, where est_names[[i]] specifies the names for est[[i]].
`extra_rows`	A character vector additional information to display for each model, such as fixed effects or controls.
`stats`	A character vector specifying which model statistics should be kept in the output. The names of the statistics should match variable names created by `glance` (r.squared, adj.r.squared, sigma, p.value). To output different summary statistics for different models, create a list of length(ms), where stats[[i]] specifies the summary statistics for ms[[i]].
`stats_names`	A character vector of labels for each object in stats. If outputting different summary statistics for different models, pass in stats_names as a list, where stats_names[[i]] specifies the names for stats[[i]].
`output_format`	A string passed to kable() that specifies the format of the table output. The options are latex, html, markdown, pandoc, and rst. The default is latex. Additionally, passing in "df" will output a dataframe version of the table, which can be used with regtable_stack() to create a table with multiple regression summaries.
`header`	A character vector to be passed into `add_header_above` that creates a new header row. This should have length equal to ms from regtable().
`sig_stars`	Logical indicating whether or not significant stars should be added to the coefficients.
`note`	A character string if a footnote is to be added to the end of the table.

# create covariates
x1 <- rnorm(1000)
x2 <- rnorm(length(x1))

# fixed effects
fe <- factor(sample(20, length(x1), replace=TRUE))

# effects for fe
fe_effs <- rnorm(nlevels(fe))

# creating left hand side y
u <- rnorm(length(x1))
y <- 2 * x1 + x2 + fe_effs[fe] + u

m1 <- felm(y ~ x1 + x2 | fe)
m2 <- glm(y ~ x1 + x2)


regtable(list(m1, m2), est = list("x1", c("x1", "x2")), 
         stats = list(c("adj.r.squared"), c("AIC")),
         stats_names = list(c("$Adj R^2$"), c("AIC")), 
         sig_stars = TRUE, output_format = "rst")