In eloualiche/FixedEffectjlr: A package for fast Fixed Effect Models in R

library(FixedEffectjlr)
JULIA_HOME <- "/Applications/Julia-0.6.app/Contents/Resources/julia/bin/"
FixedEffect_setup(JULIA_HOME)

Run the standard example from FixedEffectModels.jl

df <- Ecdat::Cigar
reg_res <- FixedEffect(df,
  lhs = "sales", rhs = "ndi",
  fe      = "state + year",
  weights = "pop",
  vcov    = "cluster(state)")

I tried to play with NSE, very much work in progress

reg_res <- FixedEffect_nse(df, sales~ndi, state+year, pop, cluster(state))

Waiting for a full lm object, I added a coeftest object so that it is possible to look at output with stargazer:

df <- Ecdat::Cigar
reg1 <- FixedEffect_nse(df, sales~ndi, state+year, pop, robust)
reg2 <- FixedEffect_nse(df, sales~ndi, state+year, pop, cluster(state))
reg3 <- FixedEffect_nse(df, sales~ndi, state+year, vcov = robust)
reg4 <- FixedEffect_nse(df, sales~ndi, state+year, vcov = cluster(state))
stargazer::stargazer(reg1$summary$coeftest, reg2$summary$coeftest,
                     reg3$summary$coeftest, reg4$summary$coeftest,
                     type = "text")

Try different models and look at standard errors:

df <- Ecdat::Cigar
reg_res <- FixedEffect_models(
  df,
  lhs = "sales", rhs = "ndi",
  fe      = "state + year",
  weights = c("", "pop"),
  vcov    = c("robust", "cluster(state)"))

eloualiche/FixedEffectjlr documentation built on April 8, 2020, 5 p.m.