Regression with plm

[Rbeginner3]

The code below runs the time series regression "return_fund ~ return market" with biglm, the function lm however didn't work.

Now I would like to switch to plm to account for fixed effects. Unfortunately plm doesn't work, like lm before.

library(biglm)
library(plm)

union_with_factors = data.table(
  t = c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5),
  excessr  = c(10,23,13,53,43,76,34,12,45,13,42,31,4,53,64),
  FundId = c("x","x","x","x","x","y","y","y","y","y","z","z","z","z","z"),
  mkt_rf = c(1,1,2,1,3,1,1,2,1,3,1,1,2,1,3)
)

sp <- split(union_with_factors, union_with_factors$FundId)
beta <- sapply(sp, function(tmp){
  fit <- plm(excessr ~ mkt_rf, data = tmp)
  coef(fit)
})```
Does anyone know what I could change that plm work ? Thanks in advance!

[clausp]

I am afraid you have to be a little more specific here. Ideally, provide a reproducible example so we have get an idea of what is not working.

FAQ: What's a reproducible example (`reprex`) and how do I create one?

Why reprex? Getting unstuck is hard. Your first step here is usually to create a reprex, or reproducible example. The goal of a reprex is to package your code, and information about your problem so that others can run it and feel your pain. Then, hopefully, folks can more easily provide a solution. What's in a Reproducible Example? Parts of a reproducible example: background information - Describe what you are trying to do. What have you already done? complete set up - include any library() calls and data to reproduce your issue. data for a reprex: Here's a discussion on setting up data for a reprex make it run - include the minimal code required to reproduce your error on the data…

[Rbeginner3]

Yes sorry! I added one.

[williaml]

Can you make your dataset reproducible, like in the article?

[Rbeginner3]

Isn't the mock code I've written a repex? My original dataset would be too big to copy it in here.

[williaml]

No. Have another look at the article. We are after the code behind your tibble. Not the output from it.

It just makes it easier for others to help you if you make it reproducible.

[Rbeginner3]

Actually the table I have created is how my data set looks like. Thats my input not the output. I would like to run the regression "excessr ~ mktrf".
Or have I understood sth. fundamentally wrong? Sorry for the confusion

[williaml]

Did you read the article?

What is the table that you have pasted in there? Do you expect someone to copy and paste it somehow in to R? You try it from your post. It won't work, that is why it needs to be reproducible. It makes it easier for other people to help you. You probably would have got a good answer hours ago.

Also, what libraries have you loaded? What is your first dataset meant to be?

[Rbeginner3]

Ok definitely my fault! Now it should be copyable.

[clausp]

You were almost there in terms of the reprex. They do take some time getting the hang off. It is, however, much easier to help if you have all the libraries there. Your reprex should have looked like this. Note, I changed your model to lm and that works fine. The output in beta is not the prettiest, but you get all the results from the three FundId values.

library(biglm)
#> Loading required package: DBI
library(plm)
library(data.table)
#> 
#> Attaching package: 'data.table'
#> The following object is masked from 'package:plm':
#> 
#>     between

union_with_factors = data.table(
  t = c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5),
  excessr  = c(10,23,13,53,43,76,34,12,45,13,42,31,4,53,64),
  FundId = c("x","x","x","x","x","y","y","y","y","y","z","z","z","z","z"),
  mkt_rf = c(1,1,2,1,3,1,1,2,1,3,1,1,2,1,3)
)

sp <- split(union_with_factors, union_with_factors$FundId)
beta <- sapply(sp, function(tmp){
  fit <- lm(excessr ~ mkt_rf, data = tmp)
  coef(fit)
})

^{Created on 2020-03-19 by the reprex package (v0.2.1)}

If you want fixed effects over FundId you cannot loop over it using sapply and you need a factor or an id number for plm to work. The following does the trick and produces a result.

library(biglm)
#> Loading required package: DBI
library(plm)
library(data.table)
#> 
#> Attaching package: 'data.table'
#> The following object is masked from 'package:plm':
#> 
#>     between

union_with_factors = data.table(
  t = c(1,2,3,4,5,1,2,3,4,5,1,2,3,4,5),
  excessr  = c(10,23,13,53,43,76,34,12,45,13,42,31,4,53,64),
  FundId = c("x","x","x","x","x","y","y","y","y","y","z","z","z","z","z"),
  mkt_rf = c(1,1,2,1,3,1,1,2,1,3,1,1,2,1,3)
)

sp <- split(union_with_factors, union_with_factors$FundId)
beta <- sapply(sp, function(tmp){
  fit <- lm(excessr ~ mkt_rf, data = tmp)
  coef(fit)
})

union_with_factors$my_fund <- factor(union_with_factors$FundId)
plm(excessr ~ mkt_rf, data = union_with_factors, model = "within", index = c("my_fund"))
#> 
#> Model Formula: excessr ~ mkt_rf
#> 
#> Coefficients:
#>  mkt_rf 
#> -4.2292

^{Created on 2020-03-19 by the reprex package (v0.2.1)}