[R] Problem with package plm

[Luca Gagliardone]

Hi!

I am trying yo run a regression using the plm package.
I tried in two different ways, that should give the same result, but
eventually do not; and i cannot understand the reason.

- The first way:
Accellerator.model <- read.delim("~/Desktop/Tesi/Accellerator model.txt")
data <- Accellerator.model
Panel <- plm.data(Accellerator.model,index=c("ID","Year"))
y <- data$Net.Cap.Inv
p  <- data$Oil.Price
rev <- data$Revenue
ebi <- data$EBITDA
tas <- data$Tot.Assets
fas <- data$Non.current
tas1 <- data$tas1
rev1 <- data$rev1
rev2 <- data$rev2
fas1 <- data$fas1
cap1 <- data$cap1
# FIXED 1
fixed <- plm(y~fas1+rev+rev1+p,data=pdata,model="within")
summary(fixed)
Oneway (individual) effect Within Model

Call:
plm(formula = y ~ fas1 + rev + rev1 + p, data = pdata, model = "within")

Unbalanced Panel: n=19, T=6-19, N=282

Residuals :
    Min.  1st Qu.   Median  3rd Qu.     Max.
-91.6000  -8.4600   0.0711   7.1000 110.0000

Coefficients :
       Estimate Std. Error t-value  Pr(>|t|)
fas1 -0.0231717  0.0096980 -2.3893  0.017595 *
rev   0.0922257  0.0035990 25.6251 < 2.2e-16 ***
rev1  0.0127055  0.0058961  2.1549  0.032093 *
p     0.1587968  0.0579552  2.7400  0.006571 **
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Total Sum of Squares:    804760
Residual Sum of Squares: 178290
R-Squared      :  0.77845
      Adj. R-Squared :  0.71496
F-statistic: 227.513 on 4 and 259 DF, p-value: < 2.22e-16
>


- The second way:
Accellerator.model <- read.delim("~/Desktop/Tesi/Accellerator model.txt")
data <- plm.data(Accellerator.model,index=c("ID","Year"))
data$y <- data$Net.Cap.Inv
data$p  <- data$Oil.Price
data$rev <- data$Revenue
data$ebi <- data$EBITDA
data$tas <- data$Tot.Assets
data$fas <- data$Non.current
Panel <- data
fixed <- plm(y~fas1+rev+rev1+p,data=Panel,model="within")
summary(fixed)
Oneway (individual) effect Within Model

Call:
plm(formula = y ~ fas1 + rev + rev1 + p, data = Panel, model = "within")

Unbalanced Panel: n=19, T=9-19, N=296

Residuals :
   Min. 1st Qu.  Median 3rd Qu.    Max.
-75.200  -3.710  -0.604   3.150  80.000

Coefficients :
      Estimate Std. Error t-value  Pr(>|t|)
fas1 0.0892306  0.0066033 13.5131 < 2.2e-16 ***
rev  0.0120747  0.0111825  1.0798    0.2812
rev1 0.0419186  0.0095914  4.3704 1.764e-05 ***
p    0.0078837  0.0375021  0.2102    0.8337
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Total Sum of Squares:    774610
Residual Sum of Squares: 85942
R-Squared      :  0.88905
      Adj. R-Squared :  0.81997
F-statistic: 546.893 on 4 and 273 DF, p-value: < 2.22e-16

As you can see, the results are totally different, and I have no
explanation for that!
Thanks for the help

Regards.
Luca.

	[[alternative HTML version deleted]]