0

在此处输入图像描述我设计了一个带有虚拟变量的 twfe 模型，以便在 diff 中进行 diff。为了引入新的自变量，我需要再引入两个虚拟变量（北/南和海/湖自治市）。如果我在代码中添加它们然后总结结果，我看不到这些系数的估计值。

不能引入多个假人吗？对于其他变量，例如人口统计数量，我没有任何问题。

代码下方：

## read the data
library(readxl)
data <- read_excel("~/link")

## load packages
library(dplyr)      ## for wrangling the data
library(plm)        ## for the regression
library(lmtest)     ## for robust standard errors

## coerce MUNICIPALITY and YEAR variables to factor
data <- data %>%
  mutate_at(c("MUNICIPALITY", "YEAR"), as.factor)

## run the regression
model_beds <- plm(formula = beds ~ DID + balneare, data = data, effect = "twoways",
                  model = "within", index = c("MUNICIPALITY", "YEAR"))

## summarize the results of the regression
summary(model_beds)

输出=

Call:
plm(formula = hotelbeds ~ DID + sea, data = data, effect = "twoways", 
    model = "within", index = c("MUNICIPALITY", "YEAR"))

Balanced Panel: n = 78, T = 18, N = 1404

Residuals:
     Min.   1st Qu.    Median   3rd Qu.      Max. 
-1229.853   -64.005   -11.272    71.209  1728.606 

Coefficients:
    Estimate Std. Error t-value Pr(>|t|)  
DID   55.471     23.342  2.3765  0.01762 *
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Total Sum of Squares:    73147000
Residual Sum of Squares: 72833000
R-Squared:      0.0042991
Adj. R-Squared: -0.068019
F-statistic: 5.64753 on 1 and 1308 DF, p-value: 0.017623

如何解决此问题并引入多个虚拟变量？

数据结构如下：

4

0 回答 0