felm function doesn't work inside lapply

问题

This is a follow-up question from

Automate regression with specific dependent and independent variables

This code works a treat with plm function from plm package. However, I wanted to estimate the same model with felm function from lfe package to obtain standard errors clustered by space (in my case county).

Let this be the data set:

regDf <- data.frame(year = rep(seq(1966,2015,1), 8), 
           county = c(rep('first', 50), rep('second', 50), rep('third', 50), rep('fourth', 50),
                      rep('fifth', 50), rep('sixth', 50), rep('seventh', 50), rep('eighth', 50)),
           crime1 = runif(400), crime2 = runif(400), crime3 = runif(400), 
           uvar1 = runif(400), uvar2 = runif(400), uvar3 = runif(400),
           var1 = runif(400), var2 = runif(400), var3 = runif(400), var4 = runif(400), var5 = runif(400))

This code works

set.seed(123)

#a helper function to create "log(var)" from "var"
fn_appendLog = function(x) {
 paste0("log(",x,")")
}



modelList = lapply(1:3,function(x) {


indepVars2 = Reduce(function(x,y) paste(x,y,sep="+"),lapply(colnames(regDF)[grepl("^v",colnames(regDF))],fn_appendLog))

#> indepVars2
#[1] "log(var1)+log(var2)+log(var3)+log(var4)+log(var5)"


indepVars1 = fn_appendLog(paste0("uvar",x))

depVar = fn_appendLog(paste0("crime",x))

formulaVar = formula(paste0(depVar, " ~ ",indepVars1,"+", indepVars2))

#> formulaVar
#log(crime1) ~ log(uvar1) + log(var1) + log(var2) + log(var3) +  log(var4) + log(var5)


modelObj = plm(formulaVar, model = 'within', effect = 'twoways', data = regDf)


})

The purpose of this code was to estimate n different models with n different unique independent variables and x common independent variables.

The new code differs only in this line:

modelObj = felm(formulaVar | county + year | 0 | county, data =regDf)

The syntax of the felm function is the following:

The first part consists of ordinary covariates, the second part consists of factors to be projected out. The third part is an IV-specification. The fourth part is a cluster specification for the standard errors.

To be concrete, these means I have a model (formulaVar) to be estimated, with county and year fixed effects, no IV (0) and clustered s.e. by county.

The following error occurs:

Error in terms(formula(as.Formula(formula), rhs = 1), specials = "G") : object 'county' not found

N.B., when estimating outside of lapply function, such as this:

felm <- felm(log(crime2) ~ log(uvar2) + log(var1) + log(var2) + log(var3) | county + year | 0 | county , data = regDf)

the function works perfectly. More on the felm function from lfe package:

https://cran.r-project.org/web/packages/lfe/lfe.pdf

来源：https://stackoverflow.com/questions/43494631/felm-function-doesnt-work-inside-lapply

标签

lapply