[英]Fixed effects regression with plm
下面的代碼使用 biglm 運行時間序列回歸“excessr ~ mkt_rf”,因為函數 lm 不適用於我的真實數據集。
現在我想從 biglm 切換到 plm 以解決固定效應。 不幸的是 plm 不起作用。
有誰知道我可以改變 plm 的工作嗎? 提前致謝!
library(biglm)
library(plm)
library(data.table)
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)
})
假設個人的 ID 由FundId
給出,時間 ID 由t
給出,以下是應用固定效應回歸的方法:
library(biglm)
library(data.table)
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)
)
fit <- plm(excessr ~ mkt_rf,
data = union_with_factors,
index = c("FundId", "t"),
model = "within")
summary(fit)
fixef(fit)
有關更多詳細信息,請參閱此處和plm
文檔(控制台中的?plm
)
編輯:在這篇文章和這篇文章之后,您似乎可以使用pmg
(也在plm
包中)進行 Fama-MacBeth 回歸:
fama_macbeth <- pmg(excessr ~ mkt_rf,
data = union_with_factors,
index = c("FundId", "t"))
summary(fama_macbeth)
聲明:本站的技術帖子網頁,遵循CC BY-SA 4.0協議,如果您需要轉載,請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.