plm具有相同规格提供两个不同的输出R

编程问答 2022-06-08

问题描述

我正在使用作者的原始数据和代码在R中复制已发布的分析。该分析基于“ plm”包中的“ plm”功能。在他们的代码中，作者在模型规范内记录了一个 变量，而在模型运行之前 进行了记录。 运行plm时，我得到两个不同的结果，对我来说，它们看起来像是相同的模型规格。

注意：在这种情况下，我试图在两个模型和结果中排除变量“ log_year”和“ log（as.numeric（Year））”。另外，两个变量“ log_year”和“ log（as.numeric（Year））”相同，因为EVDataEU $ log_year == log（as.numeric（EVDataEU $ Year））都为TRUE。

我的代码： 

log_year <- log(as.numeric(EVDataEU$Year))

formI <- log(EVRegShare) ~ MonetaryIncentive + log(ElecDieselRatio) + log_year
regI_mine = plm(formI,data = subset(EVDataEU,EVRegShare > 0),index = c("Country","Year"),model = "within",effect = "individual")
summary(regI_mine)


Oneway (individual) effect Within Model

Call:
plm(formula = formI,effect = "individual","Year"))

Unbalanced Panel: n = 32,T = 4-8,N = 226

Residuals:
     Min.   1st Qu.    Median   3rd Qu.      Max. 
-3.143446 -0.386172  0.074956  0.432507  2.446978 

Coefficients:
                        Estimate  Std. Error t-value  Pr(>|t|)    
MonetaryIncentive       0.083273    0.028787  2.8928  0.004262 ** 
log(ElecDieselRatio)   -1.652505    0.552593 -2.9905  0.003153 ** 
log_year             1227.056867   68.444862 17.9277 < 2.2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Total Sum of Squares:    481.22
Residual Sum of Squares: 126.55
R-Squared:      0.73701
Adj. R-Squared: 0.6902
F-statistic: 178.424 on 3 and 191 DF,p-value: < 2.22e-16

作者代码： 

formI <- log(EVRegShare) ~ MonetaryIncentive + log(ElecDieselRatio) + log(as.numeric(Year))
regI_CFE = plm(formI,effect = "individual")
summary(regI_CFE)


Oneway (individual) effect Within Model

Call:
plm(formula = formI,N = 226

Residuals:
      Min.    1st Qu.     Median    3rd Qu.       Max. 
-3.2890902 -0.3000665  0.0057662  0.3906836  2.4634751 

Coefficients:
                       Estimate Std. Error t-value Pr(>|t|)    
MonetaryIncentive      0.054095   0.026129  2.0703  0.03977 *  
log(ElecDieselRatio)  -0.195497   0.450552 -0.4339  0.66485    
log(as.numeric(Year))  2.224452   0.105212 21.1426  < 2e-16 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Total Sum of Squares:    481.22
Residual Sum of Squares: 101.64
R-Squared:      0.78879
Adj. R-Squared: 0.75119
F-statistic: 237.769 on 3 and 191 DF,p-value: < 2.22e-16

这么小的变化怎么可能带来如此不同的结果？

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）

logarithmplmrrspecifications