A set of estimators and tests for panel data.
Changes since version 1.6-5
Changes since version 1.6-4
Changes since version 1.6-3
Changes since version 1.6-2
Changes since version 1.6-1
Changes since version 1.5-35
Changes since version 1.5-34
Changes since version 1.5-33
Changes since version 1.5-32
Changes since version 1.5-31
Changes since version 1.5-30
Changes since version 1.5-29
Changes since version 1.5-28
Changes since version 1.5-27
Changes since version 1.5-26
Changes since version 1.5-25
Changes since version 1.5-24
Changes since version 1.5-23
Changes since version 1.5-22
Changes since version 1.5-21
Changes since version 1.5-20
Changes since version 1.5-19
Changes since version 1.5-18
Changes since version 1.5-17
* lagt.pseries: new method respecting "content" of time periods [not yet exported due to testing].
* is.pconsecutive: default method exported so it can be used for arbitrary vectors.
Changes since version 1.5-16
* plmtest: * fixed p-values [for type="kw" and "ghm"],
* unbalanced version of all test statistics implemented,
* doc update with literature references to unbalanced tests,
* if requested, the "kw" statistic is now also calculated as one-way ("individual" or "time"),
albeit they coincide with respective "bp" statistics.
* pwtest: formula interface respects 'effect' argument, 'effect' argument now mentioned in doc.
* data set 'Wages': factor 'sex' re-leveled to c("male", "female") to match use in original paper.
* print.summary.plm: suppress printing of 'effects' argument in top line in case of 'pooling' model.
* doc for between, Between, Within extended; doc for lag, lead, diff in separate file now (lag_lead_diff.Rd)
* pdata.frame: * fixed bug: do not assume a specific order of data when only individual index is supplied,
* resulting pdata.frame is ordered by individual, then time index,
* when duplicate couples (id-time) are created, a warning is issued,
* new argument 'stringAsFactors'.
* pvar: * fixed warning about var on factor variable (var on factors is deprecated as of R 3.2.3),
* fixed corner case with one group being all NA and other non-varying,
* print.pvar: better handling of NA values.
* lag/lead: fixed bug with dropped factor levels, added testfile tests/test_lag_lead_factor_levels.R.
* is.pconsecutive: * new function to check if time periods are consecutive per individual,
* better NA handling, added argument rm.na.tindex.
* pgmm: fixed bugs affecting the instrument matrix in the following cases:
* gmm instruments are collapsed and restricted in lag length;
* first lags are used as gmm instruments;
* gmm instruments are collapsed in system GMM;
* gmm instruments are restricted in lag length in system GMM.
* punbalancedness: data frame interface gains 'index' argument.
* within_intercept: new function to calculate an overall intercept along its standard error for FE models
a la Stata (and Gretl), accepts argument '.vcov' for user defined vcov.
* added help topic on package as a whole (?`plm-package`) for sake of completeness.
* summary.plm: * argument '.vcov' can also be a function (before, only only matrix was possible).
* internal: the furnished vcov is saved in the summary.plm object in object$rvcov
(vcov function name in attr(object$rvcov, which="rvcov.name").
* Ftest: * gained '.vcov' argument, which enables robust F test and chi-sq test
[robust versions not yet weaved in summary.plm].
* now exported and has documentation (.Rd file).
* returned htest object has vcov's name in 'method' element (if vcov was supplied)
* (internal) vcovXX functions: furnished vcovs gain attribute "cluster" which give info about clustering,
e.g. "group" or "time".
* fixef: gains new argument '.vcov'.
Changes since version 1.5-15
* punbalancedness: new function for unbalancedness measures for panel data as
in Ahrens/Pincus (1981); added doc and testfile for punbalancedness.
* DESCRIPTION: added URLs for package on CRAN and package on R-Forge.
* model.matrix.pFormula and plm.fit: include 'assign' and 'contrasts' attributes (if any) in model fit
* Vignette: summary(lme_mod)$coef$fixed -> summary(lme_mod)$coefficients$fixed to avoid
partial matching of 'coef' to 'coefficients' by "$" operator.
* r.squared: adjusted R squared fixed (at least for models with intercept).
* model.matrix.pFormula and pmodel.response: ensured that 'data' gets reduced to the corresponding
model frame if not already a model frame (now mimics stats::model.matrix in this regard);
fixes corner cases with specific NA patterns when model.matrix.pFormula or pmodel.response are
called directly and 'data' not being a model frame (despite being required so)
[see tests/test_model.matrix_pmodel.response_NA.R].
* detect_lin_dep: new function as a little helper function to detect linear dependent
columns, esp. in model matrices; incl. doc with two examples about how linear
dependent columns can be induced by the within transformation.
* doc pFormula.Rd extended (especially examples) and split up in two files to better accommodate
different return values and input values in the documentation
(new file added: man/model.frame_model.matrix.Rd).
Changes since version 1.5-14
* mylm: added commented (i.e. inactive) warning about dropped coefficients in estimation.
* fitted.plm: added commented (i.e. inactive) warning in about dropped coefficients in
estimated model compared to specified model.matrix.
* added testfile tests/test_fitted.plm.R (some of those test currently do not run
(commented, i.e. inactive)).
* pmodel.response.pFormula: make sure supplied formula is a pFormula before we continue
(coerce to pFormula), fixes "bugs" (rather unexpected, but documented behaviour) like:
pmodel.response.pFormula(regular_formula, data = dat, model = "pooling")
# Error in names(y) <- namesy :
# 'names' attribute [482] must be the same length as the vector [0]
* some testfiles: fixed wired encodings.
* lag.pseries: modified to handle negative lags (=leading values).
* lead.pseries: added function as a wrapper for lag.pseries(x, k=-1) for convenience,
i.e. lag(x, k = -1) == lead(x, k = 1).
* diff.pseries: prevented negative lags as input to avoid confusion.
* doc for pseries functions are made available under their name, e.g. ?lag now displays helpfile for
lag.pseries in the help overview (besides e.g. stats::lag).
* pdim.default: make error message "duplicate couples (time-id)" printed as proper error message
(removed cat() around the message in stop(), was printed as regular string on screen before).
* plm.data: slight improvement for printed outputs (spelling and spacing).
* indexes: fixed return value (was always NULL).
* doc updates: ?pdim: added section about possible different return values for pdim(pdata.frame) and
pdim(panelmodel_object); others: linkage to base functions enabled, spelling.
Changes since version 1.5-13
* fixed bug in vcovHC(..., method="white") from degenerating diag() if any groups have
only 1 element.
* vcovG framework: reintroduced "white2" method.
Changes since version 1.5-12
* dataset Produc: added variable 'region' as 3rd column and description: 'pcap' is public
capital (not private) while 'pc' is private capital (not public)
* added importFrom as per CRAN check for submission.
Changes since version 1.5-11
* added RiceFarms to datasets to eliminate non-CRAN dependencies, (temporarily) removed
'pder' from suggested, uncommented (fixed) example in pdwtest.Rd.
Changes since version 1.5-9
* fixed bug in vcovG for lagged White terms (make pseudo-diagonal in E(u,v)); affected vcovNW.
* documentation updates (mostly text books of Baltagi and Wooldridge to latest edition,
references are also more specific now).
* pbgtest: fixed/enabled type="F" (F test of lmtest::bgtest) in wrapper pbgtest() and
fixed/enabled passing of argument order.by= [additional argument in
lmtest::bgtest] [tests/testpbgtest.R added, docs updated and extended].
* phtest( , method="aux") [Hausman regression-based test]: fixed bug when data= contains more
cases than there are used in model estimation (due to NA values); avoid calc. of RE model
a second time; phtest now also works with between model correctly (fixed degrees of freedom)
tests/test_phtest_Hausman_regression.R added.
* plm(): original row names of inputted data are preserved in plm_object$model,
so functions like pmodel.response(), model.frame(), model.matrix(), residuals()
return the original row.names (and thus fancy rownames if those were to be computed by
pdata.frame); as.data.frame.pdata.frame(): respects row.names argument.
Changes since version 1.5-8
Changes since version 1.5-6
* pruned 'require' calls to 'lmtest', 'car', 'lattice' and substituted them with proper
entries in NAMESPACE.
* temporarily commented problematic examples in pbgtest and pdwtest.
Changes since version 1.5-5
* fixed bug affecting vcovG on unbalanced datasets (thx Liviu Andronic) from propagation
of NAs in final by-group matrix multiplication.
Changes since version 1.5-4
* fixed testErrors.R with plm.data instead of pdata.frame.
Changes since version 1.5-3
* reintroduced plm.data eliminated by mistake.
Changes since version 1.5-1
* fixed "already a pdata.frame" bug in pcdtest.formula.
Changes since version 1.5-0
* added generic building block and high-level wrappers vcovNW and vcovDC to the namespace.
* dataset Parity added to /data.
Changes since version 1.4-0
* substituted vcovHC, vcovSCC with the new framework based on vcovG and wrapper functions.
Changes since version 1.3-1
* a subset argument is added to print.summary.plm and summary.pht
so that a subset of coefficient may be selected.
* fixed a small problem on the printing of the typology of the
variables for pht models.
* the "name" of the tests is now the formula truncated so that it
prints on only one line.
* restrict.matrix argument added to deals with linear restrictions.
* a vcov argument is added to summary.plm so that a variance matrix can be supplied.
* the deviance method for panelmodel objects is now exported (12/02).
* the hausman-taylor now support the Amemiya MaCurdy and the Breush-Mizon-Schmidt version.
* a small bug is fixed on the var2lev function to deal the case when there are no factors.
Changes since version 1.3-0
* an update method for panelmodel objects is provided.
* the Wages example is removed from the pvar man page because it's time consuming.
Changes since version 1.2-10
* for unbalanced hausman-taylor model, the printing of the error
components was wrong, it is now fixed.
* the printing of the removed variables (cst or NA) is improved.
* the pgmm function has been improved to deal correctly with holes
in the cross-sections ; a collapse argument is added to limit the
number of gmm instruments.
* the CoefTable element of summary.pgmm objects is renamed
coefficients, so that it can easily be extracted using the coef
method.
* the default value for robust is now TRUE in the summary.pgmm
method.
* a new argument lost.ts is added in pgmm to select manually the
number of lost time series.
* almost null columns of instruments are removed (this happens
when within/between transformation is performed on between/within
series.
* plm now accept three part formulas, the last part being for
instruments introduced only in the within form.
* the predict method for panelmodel objects is now exported.
* plm now estimate systems of equations if a list of formulas is
provided as the first argument.
Changes since version 1.2-9
* the pccep function, estimating CCEP models a la Pesaran, has
been added together with summary and print.summary methods. The
function generates objects of a class of their own ("pccep"), much
like 'pggls', together with 'panelmodel'.
* the pmg function, estimating MG, DMG and CCEMG models a la
Pesaran, has been added together with summary and print.summary
methods. The function generates objects of a class of their own
("pmg"), much like 'pggls', together with 'panelmodel'. In the
future must consider possible merger with 'pvcm'.
the new cipstest function performs a second-generation CIPS test for unit roots on pseries objects.
the new (hidden) function pmerge is used internally by cipstest to merge lags and differences of a pseries into the original pdata.frame. Will possibly be added to the user space in the future.
Changes since version 1.2-8
* an index method is added for panelmodel, pdata.frame and pseries.
* a bug in the typology of the variables in pht is fixed.
* a bug in vcovBK (matrices degenerating into vectors) is fixed
(thx David Hugh-Jones for bug report).
* the Between function now returns a pseries object.
* the resid and fitted method now returns a pseries object.
* the pgmm method has been rewritten; the data frame is then first
balanced and NAs are then overwritten by 0s.
Changes since version 1.2-7
* a typo is corrected in the man page of plm.data.
* package AER is now suggested.
Changes since version 1.2-6
* a bug in mtest for pgmm models with effect=individual and
transformation=ld *and* for the wald test for time.dummies for
model with effect = twoways and transformation=ld is fixed by
modifying namest in gmm.
* there was a bug in pgmm for models with different lags for gmm
instruments. The number of time-series lost is now the min (and
not the max) of the first lags for gmm instruments.
* some parts of summary.pgmm only worked correctly for model with
time-dummies. It now deals correctly for model with 'individual'
effects.
* the rda files are now compressed.
* p-values for the two-tailed versions of plmtest() were wrong and
have been fixed. They were divided by 2 instead of multiplied.
Changes since version 1.2-5
* fixed error in pggls, model="within" (FEGLS). Added model="fd"
(FDGLS).
* changed dependency from kinship to bdsmatrix (as suggested by
Terry Therneau after his reorganization of the packages).
* fixed DESCRIPTION and NAMESPACE accordingly.
* fixed the example in pggls.Rd.
Changes since version 1.2-4
* bug corrected in pgmm : ud <- cbind(ud, td.gmm.level) is
relevant only for twoways models.
* in fitted.plm, the extraction of the index is updated.
* the residuals.plm method now has a model argument.
* a new function r.squared.
* pmodel.response.plm is modified : no explicit effect and model
arguments anymore (like in model.matrix.plm).
Changes since version 1.2-3
* lag.pseries now returns relevant names for the returned factor.
* pFormula is modified so that it can handle correctly Formula
object, and not only formula.
* pgmm has been completely rewrite with a new 'Formula'
interface. Old formula and dynformula interfaces are kept for
backward compatibility but won't be maintained in the future.
* subset, na.action are added to the list of arguments of pgmm and
oldpgmm.
* lag.pseries is now able to deal with vector arguments for lags,
e.g. lag(x, c(1,3)).
* suml(x) is replaced by Reduce("+", x).
Changes since version 1.2-2
* the documentation has been improved.
* the pvalue for purtest(..., type = "madwu") was in error (by a
factor of 2).
* in formula(dynformula), a bug is fixed so that the endog
variable doesn't appear on the rhs if there are no lags.
* in pgmm, the extract.data has been rewritten and is *much* faster.
* two new functions vcovBK and vcovSCC have been added.
* a 'model' argument is added to pgmm.sys and pgmm.diff
(previously, the model name was extracted from the call).
* Kt is fixed to 0 in pgmm when effect="individual".
Changes since version 1.2-1
* a new purtest function to perform unit root tests.
* [[.pdata.frame is modified so that NULL (and not an error
message) is returned if an unknown column is selected.
Changes since version 1.2-0
* the as.matrix and print methods for pserie objects are now
exported.
* in summary.plm, the p-value is now computed using a Student
distribution and not a normal one.
* the lag.pserie method is modified so that it deals correctly
with factors, and not only with numeric vectors. The diff.pserie
method returns an error if its argument is not numeric.
* the instruments-"backward compatibility" stuff in plm is
simplified thanks to the new features of Formula.
* 'pserie' is renamed 'pseries'.
* a THANKS file is added.
* the `[.pdata.frame` function is modified so that [, j] returns a
pseries and not a pdata.frame when j is a single integer, and a
backward compatibility feature is added for the "index" attribute.
Change since version 1-1.4
* a args argument is added for plm objects, and the internal
function relies now on it (and not on the call as previously).
* more attention is paid when one of the estimated components of
the variance is negative (warning or error messages result).
* pdata.frame objects are re-introduced. They are used to compute
model.frames. Extraction from pdata.frames results in 'pserie'
objects which have an index attribute.
* the print method of ercomp is now exported.
* the first argument of pgmm may now be a formula. A lag.form must
be provided in this case.
* hausman-taylor estimation is now performed by the pht
function. For backward compatibility reasons, it is still
possible to estimate this model with plm.
Changes since version 1-1.3
* a relevant error message is added when a within model is
estimated with no time-varying variable.
* the formula method for dynformula objects is now exported.
* a misleading notation was corrected for plm.ht model.
* the definition of sigma2$id for unbalanced ht model is
corrected, an harmonic mean of Ti being used.
* the definition of tss.default is simplified.
* the fitted.values element was missing for plm objects and has
been added.
Changes since version 1-1.2
* the inst directory was missing, it has been added again.
Changes since version 1-1.1
* part of the "details" section of the fixef.plm man page is
removed.
* a fitted.value method is now available for plm objects. It
returns the fitted values of the untransformed response.
* in pdiff, a drop=FALSE is added. This omission was responsible
of a bug while estimating a model like plm(inv~value-1, data =
Grunfeld, model = "fd").
* the lev2var is changed so that it doesn't result in an error
when the data.frame contains no factor: this was responsible for a
bug in plm.ht.
Changes since version 1-1.0
* in fixef, the effect argument default is now NULL:
fixef(model_with_time_effect) now works correctly.
* in pFtest, the error message "the arguments should be a within
and a pooling model" is removed so that two within models may be
provided.
* for backward compatibility reasons, the pvcovHC is reintroduced.
* for backward compatibility reasons, argument test of pbsytest
may be indicated in upper case.
* we switched back to old names for two arguments of plm ercomp ->
random.methods ivar -> inst.method -> ivar.
* amemiya method is not implemented for unbalanced panels: an
error message is now returned.
Changes since version 1-0.1
* the plm function has been completely rewritten.
* the names of some arguments have changed (random.methods ->
ercomp, inst.method -> ivar), the old names are accepted with a
warning.
* the instruments argument is removed, instrumental variable
estimation is performed using extended formula. The instruments
argument is still accepted with a warning.
* the model argument of plm objects are now ordinary data.frame,
and not data.frame with elements y and X. Moreover, this
data.frame contains untransformed data.
* the data sets which are relevant for panel data estimation that
where previously in the Ecdat package are now in the plm package.
* in pvcm a bug when the estimation was made on a subset is fixed.
* ercomp is a stand alone function which returns the estimation of
the components of the variance of the errors.
* the estimation of two-ways within model is now implemented for
unbalanced panels.
* the fixef method is much improved, the fixed effects may be
computed in levels, in deviation from the first level or in
deviation from the overall mean.
* in pbsytest, the arguments test are now in lowercase.
* the pvcovHC function is replaced by suitable vcovHC methods for
panelmodel and pgmm models.
Changes since version 1-0.0
* lag and diff methods for pseries are now exported and therefore
behave correctly.
* for two-ways within models with instrumental variables, K is
replaced by K+1 for the computation of the overall mean matrix of
the instruments. Time fixef are now computed. The error message
"impossible ..." is removed.
* a bug in the time random effect model is corrected.
* a model.matrix method for panelmodel is provided.
* models without intercept (-1 in the formula) should now be
consistently estimated with plm, pggls and pvcm.
* plm depends now on the Formula package which provides useful
tools for formula with two parts.
Changes since version 0-3.2
* a lot of typos of the man pages have been fixed.
* functions pcdtest, pcdres have been added.
* for hausman taylor model, the summary now prints the variables
and not the effects.
* the estimation of a model with only one explanatory variable
using plm with method = "fd" is now working correctly.
Changes since version 0-3.1
* in plm.formula, [int.row.names] is replaced by
[as.character(int.row.names)].
* the degree of freedom for a within time effect model was wrong
and has been corrected.
* the arguments type and weights in pvcovHC.panelmodel are renamed
method and type respectively. The default method (type in previous
versions) is arellano and not white1.
* honda is now the default option for plmtest.
Changes since version 0-2.2
* the coefficients method for objects of class pgmm is now
exported.
* a bug in the plm method of plmtest has been fixed
* in plmtest, the argument effect "id" is renamed "individual".
* three testing functions are added : pbsytest (Bera,
Sosa-Escudero and Yoon test), pARtest (Breusch-Godfrey Test) and
pDWtest (Durbin-Watson test), pwartest, pBGtest, pwtest pbltest.
* plm, pvcm and pggls now have arguments "subset" and "na.action".
* phtest, pFtest, plmtest now have a formula method.
* a bug is fixed for the vcov of the within model.
* the pdata.frame function and class is suppressed. The package
now use ordinary data.frames.
Changes since version 0-2.1
* pdata.frame is much faster thanks to a modification of the pvar
function.
* series with only NA values or with constants values are now
removed by pdata.frame.
* observations are ordered by id and time by pdata.frame.
* a pfix function is added to edit a pdata.frame.
* a as.data.frame function is provided to coerce a pdata.frame to
a data.frame.
* the dependency to the Matrix package has been removed and pgmm
is much faster now.
* phtest has been fixed to return only positive values of the
statistic.
* pgmm objects now inherits from panelmodel like other estimators
and prints correctly.
* a bug in summary.pgmm has been fixed.
Changes since version 0-1.2
* Models with only one explanatory variable resulted in a
bug. This has been fixed.
* Estimation methods are now available with four functions : plm,
pvcm, pggls and pgmm instead of one (plm) in the previous version.
* pvcm is a new function which estimate variable coefficients
models. The "nopool" model is now part of it.
* pggls is a new function which enables the estimation of general
FGLS.
* pgmm is a new function for general method of moments estimator.
* for all estimation functions, the first four arguments are now
formula, data, effect, model.
* robust inference is now provided by the pvcovHC function.