1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/test_serial.R
\name{pwfdtest}
\alias{pwfdtest}
\alias{pwfdtest.formula}
\alias{pwfdtest.panelmodel}
\title{Wooldridge first--difference--based test for AR(1) errors in levels
or first--differenced panel models}
\usage{
pwfdtest(x, ...)
\method{pwfdtest}{formula}(x, data, ..., h0 = c("fd", "fe"))
\method{pwfdtest}{panelmodel}(x, ..., h0 = c("fd", "fe"))
}
\arguments{
\item{x}{an object of class \code{formula} or a \code{"fd"}-model (plm
object),}
\item{\dots}{further arguments to be passed on to \code{vcovHC} (see Details
and Examples).}
\item{data}{a \code{data.frame},}
\item{h0}{the null hypothesis: one of \code{"fd"}, \code{"fe"},}
}
\value{
An object of class \code{"htest"}.
}
\description{
First--differencing--based test of serial correlation for (the idiosyncratic
component of) the errors in either levels or first--differenced panel
models.
}
\details{
As \insertCite{WOOL:10;textual}{plm}, Sec. 10.6.3 observes, if the
idiosyncratic errors in the model in levels are uncorrelated (which
we label hypothesis \code{"fe"}), then the errors of the model in first
differences (FD) must be serially correlated with
\eqn{cor(\hat{e}_{it}, \hat{e}_{is}) = -0.5} for each \eqn{t,s}. If
on the contrary the levels model's errors are a random walk, then
there must be no serial correlation in the FD errors (hypothesis
\code{"fd"}). Both the fixed effects (FE) and the first--differenced
(FD) estimators remain consistent under either assumption, but the
relative efficiency changes: FE is more efficient under \code{"fe"}, FD
under \code{"fd"}.
Wooldridge (ibid.) suggests basing a test for either hypothesis on
a pooled regression of FD residuals on their first lag:
\eqn{\hat{e}_{i,t}=\alpha + \rho \hat{e}_{i,t-1} +
\eta_{i,t}}. Rejecting the restriction \eqn{\rho = -0.5} makes us
conclude against the null of no serial correlation in errors of the
levels equation (\code{"fe"}). The null hypothesis of no serial
correlation in differenced errors (\code{"fd"}) is tested in a similar
way, but based on the zero restriction on \eqn{\rho} (\eqn{\rho =
0}). Rejecting \code{"fe"} favours the use of the first--differences
estimator and the contrary, although it is possible that both be
rejected.
\code{pwfdtest} estimates the \code{fd} model (or takes an \code{fd} model as
input for the panelmodel interface) and retrieves its residuals,
then estimates an AR(1) \code{pooling} model on them. The test statistic
is obtained by applying a F test to the latter model to test the
relevant restriction on \eqn{\rho}, setting the covariance matrix
to \code{vcovHC} with the option \code{method="arellano"} to control for
serial correlation.
Unlike the \code{pbgtest} and \code{pdwtest}, this test does not rely on
large--T asymptotics and has therefore good properties in ''short''
panels. Furthermore, it is robust to general
heteroskedasticity. The \code{"fe"} version can be used to test for
error autocorrelation regardless of whether the maintained
specification has fixed or random effects
\insertCite{@see @DRUK:03}{plm}.
}
\examples{
data("EmplUK" , package = "plm")
pwfdtest(log(emp) ~ log(wage) + log(capital), data = EmplUK)
pwfdtest(log(emp) ~ log(wage) + log(capital), data = EmplUK, h0 = "fe")
# pass argument 'type' to vcovHC used in test
pwfdtest(log(emp) ~ log(wage) + log(capital), data = EmplUK, type = "HC3", h0 = "fe")
# same with panelmodel interface
mod <- plm(log(emp) ~ log(wage) + log(capital), data = EmplUK, model = "fd")
pwfdtest(mod)
pwfdtest(mod, h0 = "fe")
pwfdtest(mod, type = "HC3", h0 = "fe")
}
\references{
\insertRef{DRUK:03}{plm}
\insertRef{WOOL:02}{plm}
Sec. 10.6.3, pp. 282--283.
\insertRef{WOOL:10}{plm}
Sec. 10.6.3, pp. 319--320
}
\seealso{
\code{pdwtest}, \code{pbgtest}, \code{pwartest},
}
\author{
Giovanni Millo
}
\keyword{htest}
|
