This document contains answers to some of the most frequently asked
questions about R.
2 R Basics
2.1 What is R?
R is a system for statistical computation and graphics. It consists of
a language plus a run-time environment with graphics, a debugger, access
to certain system functions, and the ability to run programs stored in
script files.
The design of R has been heavily influenced by two existing languages:
Becker, Chambers & Wilks’ S (see
What is S?
) and Sussman’s
Scheme
Whereas the resulting language is very similar in appearance to S, the
underlying implementation and semantics are derived from Scheme.
See
What are the differences between R and S?
, for further details.
The core of R is an interpreted computer language which allows branching
and looping as well as modular programming using functions. Most of the
user-visible functions in R are written in R. It is possible for the
user to interface to procedures written in the C, C++, or FORTRAN
languages for efficiency. The R distribution contains functionality for
a large number of statistical procedures. Among these are: linear and
generalized linear models, nonlinear regression models, time series
analysis, classical parametric and nonparametric tests, clustering and
smoothing. There is also a large set of functions which provide a
flexible graphical environment for creating various kinds of data
presentations. Additional modules (“add-on packages”) are available
for a variety of specific purposes (see
R Add-On Packages
).
R was initially written by
Ross Ihaka
and
Robert Gentleman
at the
Department of Statistics of the University of Auckland in Auckland, New
Zealand. In addition, a large group of individuals has contributed to R
by sending code and bug reports.
Since mid-1997 there has been a core group (the “R Core Team”) who can
modify the R source code archive, currently consisting of
John Chambers,
Peter Dalgaard,
Robert Gentleman,
Kurt Hornik,
Ross Ihaka,
Tomas Kalibera,
Michael Lawrence,
Uwe Ligges,
Thomas Lumley,
Martin Maechler,
Sebastian Meyer,
Paul Murrell,
Martyn Plummer,
Brian Ripley,
Deepayan Sarkar,
Duncan Temple Lang,
Luke Tierney, and
Simon Urbanek,
plus
Heiner Schwarte up to October 1999,
Guido Masarotto up to June 2003,
Stefano Iacus up to July 2014,
Seth Falcon up to August 2015,
Duncan Murdoch up to September 2017,
Martin Morgan up to June 2021,
Douglas Bates up to March 2024, and
Friedrich Leisch up to April 2024.
R has a home page at
. It is
free software
distributed under a
GNU
-style
copyleft
, and an
official part of the
GNU
project
(“
GNU
S”).
2.2 What machines does R run on?
R is being developed for the Unix-like, Windows and Mac families of
operating systems. Support for Mac OS Classic ended with R 1.7.1.
The current version of R will configure and build under a number of
common Unix-like (e.g.,
platforms including
cpu
-linux-gnu for the i386,
amd64/x86_64, alpha, arm, arm64, hppa,
mips/mipsel, powerpc, s390x and sparc
CPU
s (e.g.,
),
386-hurd-gnu,
cpu
-kfreebsd-gnu for i386 and amd64,
i386-pc-solaris,
rs6000-ibm-aix,
sparc-sun-solaris,
x86_64-apple-darwin,
aarch64-apple-darwin,
x86_64-unknown-freebsd and
x86_64-unknown-openbsd.
If you know about other platforms, please drop us a note.
2.3 What is the current version of R?
R uses a ‘major.minor.patchlevel’ numbering scheme. Based on this,
there are the current release version of R (‘r-release’) as well as two
development versions of R, a patched version of the current release
(‘r-patched’) and one working towards the next minor or eventually major
(‘r-devel’) releases of R, respectively. New features are typically
introduced in r-devel, while r-patched is for bug fixes mostly.
See
for the current
versions of r-release, r-patched and r-devel.
2.5 How can R be installed?
2.5.1 How can R be installed (Unix-like)
If R is already installed, it can be started by typing
at the
shell prompt (of course, provided that the executable is in your path).
If binaries are available for your platform (see
Are there Unix-like binaries for R?
), you can use these, following the
instructions that come with them.
Otherwise, you can compile and install R yourself, which can be done
very easily under a number of common Unix-like platforms (see
What machines does R run on?
). The file
INSTALL
that comes with the
R distribution contains a brief introduction, and the “R Installation
and Administration” guide (see
What documentation exists for R?
has full details.
Note that you need a FORTRAN 90 compiler as well as a C compiler to
build R.
In the simplest case, untar the R source code, change to the directory
thus created, and issue the following commands (at the shell prompt):
If these commands execute successfully, the R binary and a shell script
front-end called
are created and copied to the
bin
directory. You can copy the script to a place where users can invoke
it, for example to
/usr/local/bin
. In addition,
HTML
versions of the R manuals (e.g.,
R-exts.html
, the
“Writing R Extensions” manual) are built in the
doc/manual
subdirectory (if a suitable
texi2any
program was found).
Use
make pdf
to build PDF (Portable Document Format)
versions of the R manuals, including
fullrefman.pdf
(an R object reference index). Manuals written in the
GNU
Texinfo system can also be converted to
.info
files
suitable for reading online with Emacs or stand-alone
GNU
Info; use
make info
to create these files.
Finally, use
make check
to find out whether your R system works
correctly.
You can also perform a “system-wide” installation using
make
install
. By default, this will install to the following directories:
${prefix}/bin
the front-end shell script
${prefix}/man/man1
the man page
${prefix}/lib/R
all the rest (libraries, on-line help system, …). This is the “R
Home Directory” (
R_HOME
) of the installed system.
In the above,
prefix
is determined during configuration
(typically
/usr/local
) and can be set by running
configure
with the option
$ ./configure --prefix=/where/you/want/R/to/go
(E.g., the R executable will then be installed into
/where/you/want/R/to/go/bin
.)
To install info and PDF versions of the manuals, use
make install-info
and
make install-pdf
respectively.
2.5.2 How can R be installed (Windows)
The
bin/windows
directory of a
CRAN
site contains binaries for
a base distribution and add-on packages from
CRAN
to run on 64-bit
versions of Windows 10 and later on x86_64 chips (R 4.1.3 was the last
version of R to support 32-bit Windows). The Windows version of R was
created by Robert Gentleman and Guido Masarotto; Brian
Ripley and Duncan Murdoch made substantial contributions and it is
now being maintained by other members of the R Core team.
The same directory has links to snapshots of the r-patched and r-devel
versions of R.
See the
“R for Windows
FAQ
for more details.
2.5.3 How can R be installed (Mac)
The
bin/macosx
directory of a
CRAN
site contains a standard
Apple installer package to run on macOS 10.13 (‘High Sierra’) or later,
and another which runs only on ‘Apple Silicon’ Macs under macOS 11
(‘Big Sur’) or later. Once downloaded and executed, the installer will
install the current release of R and R.app, the macOS
GUI
This port of R for macOS is maintained by
Simon Urbanek
(and previously by
Stefano Iacus). The
“R for
macOS
FAQ
has more details.
Snapshots of the r-patched and r-devel versions of R are available as
Apple installer packages at
2.6 Are there Unix-like binaries for R?
Binary distributions of R are available on many Unix-like OSes: only
some can be mentioned here so check your OS’s search facilities to see
if one is available for yours.
The
bin/linux
directory of a
CRAN
site contains R packages for
Debian and Ubuntu.
Debian packages, maintained by Dirk Eddelbuettel, have long been part of
the Debian distribution, and can be accessed through APT, the Debian
package maintenance tool. Use e.g.
apt-get install r-base
r-recommended
to install the R environment and recommended packages.
If you also want to build R packages from source, also run
apt-get
install r-base-dev
to obtain the additional tools required for this.
So-called “backports” of the current R packages for at least the
stable
distribution of Debian are provided by Johannes Ranke, and
available from
CRAN
. See
for details on R
Debian packages and installing the backports, which should also be
suitable for other Debian derivatives. Native backports for Ubuntu are
provided by Michael Rutter, see
for
instructions.
R binaries for Fedora, maintained by Tom “Spot” Callaway and
Iñaki Ucar, are provided as part of the Fedora distribution and can
be accessed through
dnf
, the
RPM
installer/updater.
The Fedora R
RPM
is a “meta-package” which installs all the
user and developer components of R (available separately as
R-core
and
R-core-devel
), as well as
R-java
and
R-java-devel
, which ensures that R is configured for use with
Java. The R
RPM
also installs the standalone R math library
libRmath
and
libRmath-devel
), although this is not
necessary to use R. When a new version of R is released, there may be a
delay of up to 2 weeks until the Fedora
RPM
becomes publicly
available, as it must pass through the Fedora update process.
The Extra Packages for Enterprise Linux (
EPEL
) project
) provides ports of the
Fedora
RPM
s for RedHat Enterprise Linux and compatible
distributions (e.g., CentOS Stream, Scientific Linux, Oracle
Linux, AlmaLinux, or Rocky Linux among others).
RPM
s for selection of R packages are also provided by Fedora.
Additional
RPM
s for R packages are maintained by Iñaki Ucar
on Fedora Copr. See
for further details and installation instructions.
No other binary distributions are currently publicly available via
CRAN
2.7 What documentation exists for R?
Online documentation for most of the functions and variables in R
exists, and can be printed on-screen by typing
help(
name
(or
name
) at the R prompt, where
name
is the name of
the topic help is sought for. (In the case of unary and binary
operators and control-flow special forms, the name may need to be
quoted.)
This documentation can also be made available as one reference manual
for on-line reading in
HTML
and PDF formats, and as hardcopy via
LaTeX, see
How can R be installed?
. An up-to-date
HTML
version is always available for web browsing at
The R distribution also comes with the following manuals.
“An Introduction to R” (
R-intro
includes information on data types, programming elements, statistical
modeling and graphics. This document is based on the “Notes on
S-PLUS
” by Bill Venables and David Smith.
“Writing R Extensions” (
R-exts
currently describes the process of creating R add-on packages, writing R
documentation, R’s system and foreign language interfaces, and the R
API
“R Data Import/Export” (
R-data
is a guide to importing and exporting data to and from R.
“The R Language Definition” (
R-lang
),
a first version of the “Kernighan & Ritchie of R”, explains
evaluation, parsing, object oriented programming, computing on the
language, and so forth.
“R Installation and Administration” (
R-admin
).
“R Internals” (
R-ints
is a guide to R’s internal structures.
(Added in R 2.4.0.)
An annotated bibliography (BibTeX format) of R-related publications
can be found at
Books on R by R Core Team members include
John M. Chambers (2008),
“Software for Data Analysis: Programming with R”.
Springer, New York, ISBN 978-0-387-75935-7,
Peter Dalgaard (2008),
“Introductory Statistics with R”,
2nd edition.
Springer, ISBN 978-0-387-79053-4,
Robert Gentleman (2008),
“R Programming for Bioinformatics”.
Chapman & Hall/CRC, Boca Raton, FL, ISBN 978-1-420-06367-7,
Stefano M. Iacus (2008),
“Simulation and Inference for Stochastic Differential Equations: With R
Examples”.
Springer, New York, ISBN 978-0-387-75838-1.
Deepayan Sarkar (2007),
“Lattice: Multivariate Data Visualization with R”.
Springer, New York, ISBN 978-0-387-75968-5.
W. John Braun and Duncan J. Murdoch (2007),
“A First Course in Statistical Programming with R”.
Cambridge University Press, Cambridge, ISBN 978-0521872652.
P. Murrell (2005), “R Graphics”,
Chapman & Hall/CRC, ISBN 1-584-88486-X,
William N. Venables and Brian D. Ripley (2002),
“Modern Applied Statistics with S” (4th edition).
Springer, ISBN 0-387-95457-0,
Jose C. Pinheiro and Douglas M. Bates (2000),
“Mixed-Effects Models in S and S-Plus”.
Springer, ISBN 0-387-98957-0.
Last, but not least, Ross’ and Robert’s experience in designing and
implementing R is described in Ihaka & Gentleman (1996),
“R: A Language for Data Analysis and Graphics”,
Journal of Computational and Graphical Statistics
, 299–314
(doi:
10.1080/10618600.1996.10474713
).
2.8 Citing R
To cite R in publications, use
@Manual{,
title = {R: A Language and Environment for Statistical
Computing},
author = {{R Core Team}},
organization = {R Foundation for Statistical Computing},
address = {Vienna, Austria},
year =
YEAR
url = {https://www.R-project.org}
where
YEAR
is the release year of the version of R used and can
determined as
R.version$year
Citation strings (or BibTeX entries) for R and R packages can also be
obtained by
citation()
2.9 What mailing lists exist for R?
Thanks to
Martin Maechler
, there
are several mailing lists devoted to R, including the following:
R-announce
A moderated list for major announcements about the development of R and
the availability of new code.
R-packages
A moderated list for announcements on the availability of new or
enhanced contributed packages.
R-help
The ‘main’ R mailing list, for discussion about problems and solutions
encountered using R, including using R packages in the standard R distribution
and on CRAN; announcements (not covered by ‘R-announce’ or ‘R-packages’);
the availability of new functionality for R and documentation of R;
and for posting nice examples and benchmarks.
R-devel
This list is for questions and discussion about code development in R.
R-package-devel
A list which provides a forum for those developing R packages.
Please read the
posting guide
before
sending anything to any mailing list.
Note in particular that R-help is intended to be comprehensible to
people who want to use R to solve problems but who are not necessarily
interested in or knowledgeable about programming. Questions likely to
prompt discussion unintelligible to non-programmers (e.g., questions
involving C or C++) should go to R-devel.
Convenient access to information on these lists, subscription, and
archives is provided by the web interface at
. One can also subscribe
(or unsubscribe) via email, e.g. to R-help by sending ‘
(or ‘
unsubscribe
’) in the
body
of the message (not in the
subject!) to
R-help-request@lists.R-project.org
Send email to
R-help@lists.R-project.org
to send a message to
everyone on the R-help mailing list. Subscription and posting to the
other lists is done analogously, with ‘
R-help
’ replaced by
R-announce
’, ‘
R-packages
’, and ‘
R-devel
’, respectively.
Note that the R-announce and R-packages lists are gatewayed into R-help.
Hence, you should subscribe to either of them only in case you are not
subscribed to R-help.
It is recommended that you send mail to R-help rather than only to the R
Core developers (who are also subscribed to the list, of course). This
may save them precious time they can use for constantly improving R, and
will typically also result in much quicker feedback for yourself.
Of course, in the case of bug reports it would be very helpful to have
code which reliably reproduces the problem. Also, make sure that you
include information on the system and version of R being used. See
R Bugs
for more details.
See
for more information on
the R mailing lists.
2.10 What is
CRAN
The “Comprehensive R Archive Network” (
CRAN
) is a collection of
sites which carry identical material, consisting of the R
distribution(s), the contributed extensions, documentation for R, and
binaries.
The
CRAN
main site at WU (Wirtschaftsuniversität Wien) in Austria
can be found at the
URL
and is mirrored daily to many sites around the world.
See
for a complete list of
mirrors. Please use the
CRAN
site closest to you to reduce network
load.
From
CRAN
, you can obtain the latest official release of R, daily
snapshots of R (copies of the current source trees), as gzipped and
bzipped tar files, a wealth of additional contributed code, as well as
prebuilt binaries for various operating systems (Linux, Mac OS Classic,
macOS, and MS Windows).
CRAN
also provides access to
documentation on R, existing mailing lists and the R Bug Tracking
system.
Since March 2016, “old” material is made available from a central
CRAN
archive server (
).
Please always use the
URL
of the master site when referring to
CRAN
2.11 Can I use R for commercial purposes?
R is released under the
GNU
General Public License (
GPL
), version 2 or
version 3
. If you have any questions regarding the legality of using R
in any particular situation you should bring it up with your legal
counsel. We are in no position to offer legal advice.
It is the opinion of the R Core Team that one can use R for commercial
purposes (e.g., in business or in consulting). The
GPL
, like
all Open Source licenses, permits all and any use of the package. It
only restricts distribution of R or of other programs containing code
from R. This is made clear in clause 6 (“No Discrimination Against
Fields of Endeavor”) of the
Open Source Definition
The license must not restrict anyone from making use of the program in a
specific field of endeavor. For example, it may not restrict the
program from being used in a business, or from being used for genetic
research.
It is also explicitly stated in clause 0 of the GPL, which says in part
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of running
the Program is not restricted, and the output from the Program is
covered only if its contents constitute a work based on the Program.
Most add-on packages, including all recommended ones, also explicitly
allow commercial use in this way. A few packages are restricted to
“non-commercial use”; you should contact the author to clarify whether
these may be used or seek the advice of your legal counsel.
None of the discussion in this section constitutes legal advice. The R
Core Team does not provide legal advice under any circumstances.
2.12 Why is R named R?
The name is partly based on the (first) names of the first two R authors
(Robert Gentleman and Ross Ihaka), and partly a play on the name of the
Bell Labs language ‘S’ (see
What is S?
).
2.13 What is the R Foundation?
The R Foundation is a not for profit organization working in the public
interest. It was founded by the members of the R Core Team in order to
provide support for the R project and other innovations in statistical
computing, provide a reference point for individuals, institutions or
commercial enterprises that want to support or interact with the R
development community, and to hold and administer the copyright of R
software and documentation. See
for more information.
2.14 What is R-Forge?
R-Forge (
) offers a central platform
for the development of R packages, R-related software and further
projects. It is based on
GForge
offering
easy access to the best in SVN, daily built and checked packages,
mailing lists, bug tracking, message boards/forums, site hosting,
permanent file archival, full backups, and total web-based
administration. For more information, see the R-Forge web page and
Stefan Theußl and Achim Zeileis (2009), “Collaborative software
development using R-Forge”,
The R Journal
(1):9–14.
6 R and Emacs
6.1 Is there Emacs support for R?
There is an Emacs package called
ESS
(“Emacs Speaks
Statistics”) which provides a standard interface between statistical
programs and statistical processes. It is intended to provide
assistance for interactive statistical programming and data analysis.
Languages supported include: S dialects (R, S 3/4, and
S-PLUS
3.x/4.x/5.x/6.x/7.x), LispStat dialects (XLispStat, ViSta),
SAS, Stata, and BUGS.
ESS
grew out of the need for bug fixes and extensions to
S-mode 4.8 (which was a
GNU
Emacs interface to S/
S-PLUS
version 3 only). The current set of developers desired support for
XEmacs, R, S4, and MS Windows. In addition, with new modes being
developed for R, Stata, and SAS, it was felt that a unifying interface
and framework for the user interface would benefit both the user and the
developer, by helping both groups conform to standard Emacs usage. The
end result is an increase in efficiency for statistical programming and
data analysis, over the usual tools.
R support contains code for editing R source code (syntactic indentation
and highlighting of source code, partial evaluations of code, loading
and error-checking of code, and source code revision maintenance) and
documentation (syntactic indentation and highlighting of source code,
sending examples to running
ESS
process, and previewing),
interacting with an inferior R process from within Emacs (command-line
editing, searchable command history, command-line completion of R object
and file names, quick access to object and search lists, transcript
recording, and an interface to the help system), and transcript
manipulation (recording and saving transcript files, manipulating and
editing saved transcripts, and re-evaluating commands from transcript
files).
The latest stable version of
ESS
is available via
CRAN
or
the
ESS web page
ESS
comes with detailed installation instructions.
For help with
ESS
, send email to
ESS-help@r-project.org
Please send bug reports and suggestions on
ESS
to
ESS-bugs@r-project.org
. The easiest way to do this from is
within Emacs by typing
M-x ess-submit-bug-report
or using the
[ESS] or [iESS] pulldown menus.
6.2 Should I run R from within Emacs?
Yes, instead of just running it in a console,
definitely
. As an
alternative to other
IDE
s such as RStudio,
possibly
, notably if you
are interested to use Emacs for other computer interaction.
You’d be using
ESS
, Emacs Speaks Statistics, see previous FAQ.
Inferior R mode provides a readline/history
mechanism, object name completion, and syntax-based highlighting of the
interaction buffer using Font Lock mode, as well as a very convenient
interface to the R help system.
Of course, it also integrates nicely with the mechanisms for editing R
source using Emacs. One can write code in one Emacs buffer and send
whole or parts of it for execution to R; this is helpful for both data
analysis and programming. One can also seamlessly integrate with a
revision control system, in order to maintain a log of changes in your
programs and data, as well as to allow for the retrieval of past
versions of the code.
In addition, it allows you to keep a record of your session, which can
also be used for error recovery through the use of the transcript mode.
To specify command line arguments for the inferior R process, use
C-u M-x R
for starting R.
6.3 Debugging R from within Emacs
To debug R “from within Emacs”, there are several possibilities. To
use the Emacs
GUD
(Grand Unified Debugger) library with the recommended
debugger GDB, type
M-x gdb
and give the path to the R
binary
as argument. At the
gdb
prompt, set
R_HOME
and other environment variables as needed (using e.g.
set env R_HOME /path/to/R/
, but see also below), and start the
binary with the desired arguments (e.g.,
run --quiet
).
If you have
ESS
, you can do
C-u M-x R
RET
- d
SPC
g d b
RET
to start an inferior R process with arguments
-d gdb
A third option is to start an inferior R process via
ESS
M-x R
) and then start
GUD
M-x gdb
) giving the R binary
(using its full path name) as the program to debug. Use the program
ps
to find the process number of the currently running R
process then use the
attach
command in GDB to attach it to that
process. One advantage of this method is that you have separate
*R*
and
*gud-gdb*
windows. Within the
*R*
window
you have all the
ESS
facilities, such as object-name
completion, that we know and love.
When using
GUD
mode for debugging from within Emacs, you may find it
most convenient to use the directory with your code in it as the current
working directory and then make a symbolic link from that directory to
the R binary. That way
.gdbinit
can stay in the directory with
the code and be used to set up the environment and the search paths for
the source, e.g. as follows:
set env R_HOME /opt/R
set env R_PAPERSIZE letter
set env R_PRINTCMD lpr
dir /opt/R/src/appl
dir /opt/R/src/main
dir /opt/R/src/nmath
dir /opt/R/src/unix
7 R Miscellanea
7.1 How can I set components of a list to NULL?
You can use
to set component
of the list
to
NULL
, similarly
for named components. Do not set
x[i]
or
x[[i]]
to
NULL
, because this will remove the corresponding component from
the list.
For dropping the row names of a matrix
, it may be easier to use
rownames(x) <- NULL
, similarly for column names.
7.2 How can I save my workspace?
save.image()
saves the objects in the user’s
.GlobalEnv
to
the file
.RData
in the R startup directory. (This is also what
happens after
q("yes")
.) Using
save.image(
file
one
can save the image under a different name.
7.3 How can I clean up my workspace?
To remove all objects in the currently active environment (typically
.GlobalEnv
), you can do
rm(list = ls(all.names = TRUE))
(Without
all = TRUE
, only the objects with names not starting
with a ‘
’ are removed.)
7.4 How can I get
eval()
and
D()
to work?
Strange things will happen if you use
eval(print(x), envir = e)
or
D(x^2, "x")
. The first one will either tell you that
" is not found, or print the value of the wrong
The other one will likely return zero if
exists, and an error
otherwise.
This is because in both cases, the first argument is evaluated in the
calling environment first. The result (which should be an object of
mode
"expression"
or
"call"
) is then evaluated or
differentiated. What you (most likely) really want is obtained by
“quoting” the first argument upon surrounding it with
expression()
. For example,
R> D(expression(x^2), "x")
2 * x
Although this behavior may initially seem to be rather strange, it is
perfectly logical. The “intuitive” behavior could easily be
implemented, but problems would arise whenever the expression is
contained in a variable, passed as a parameter, or is the result of a
function call. Consider for instance the semantics in cases like
D2 <- function(e, n) D(D(e, n), n)
or
g <- function(y) eval(substitute(y), sys.frame(sys.parent(n = 2)))
g(a * b)
See the help page for
deriv()
for more examples.
7.5 Why do my matrices lose dimensions?
When a matrix with a single row or column is created by a subscripting
operation, e.g.,
row <- mat[2, ]
, it is by default turned into a
vector. In a similar way if an array with dimension, say, 2 x 3 x 1 x 4 is created by subscripting it will be coerced into a 2 x 3 x 4
array, losing the unnecessary dimension. After much discussion this has
been determined to be a
feature
To prevent this happening, add the option
drop = FALSE
to the
subscripting. For example,
rowmatrix <- mat[2, , drop = FALSE] #
creates a row matrix
colmatrix <- mat[, 2, drop = FALSE] #
creates a column matrix
a <- b[1, 1, 1, drop = FALSE] #
creates a 1 x 1 x 1 array
The
drop = FALSE
option should be used defensively when
programming. For example, the statement
will return a vector rather than a matrix if
index
happens to
have length 1, causing errors later in the code. It should probably be
rewritten as
somerows <- mat[index, , drop = FALSE]
7.6 How does autoloading work?
Autoloading is rarely used since packages became lazy-loaded.
R has a special environment called
.AutoloadEnv
. Using
autoload(
name
pkg
, where
name
and
pkg
are strings giving the names of an object and the package
containing it, stores some information in this environment. When R
tries to evaluate
name
, it loads the corresponding package
pkg
and reevaluates
name
in the new package’s
environment.
Using this mechanism makes R behave as if the package was loaded, but
does not occupy memory (yet).
See the help page for
autoload()
for a very nice example.
7.7 How should I set options?
The function
options()
allows setting and examining a variety of
global “options” which affect the way in which R computes and displays
its results. The variable
.Options
holds the current values of
these options, but should never directly be assigned to unless you want
to drive yourself crazy—simply pretend that it is a “read-only”
variable.
For example, given
test1 <- function(x = pi, dig = 3) {
oo <- options(digits = dig); on.exit(options(oo));
cat(.Options$digits, x, "\n")
test2 <- function(x = pi, dig = 3) {
.Options$digits <- dig
cat(.Options$digits, x, "\n")
we obtain:
R> test1()
3 3.14
R> test2()
3 3.141593
What is really used is the
global
value of
.Options
, and
using
options(OPT = VAL)
correctly updates it. Local copies of
.Options
, either in
.GlobalEnv
or in a function
environment (frame), are just silently disregarded.
7.8 How do file names work in Windows?
As R uses C-style string handling, ‘
’ is treated as an escape
character, so that for example one can enter a newline as ‘
\n
’.
When you really need a ‘
’, you have to escape it with another
’.
Thus, in filenames use something like
"c:\\data\\money.dat"
. You
can also replace ‘
’ by ‘
’ (
"c:/data/money.dat"
).
7.9 Why does plotting give a color allocation error?
This is about a problem rarely seen with modern X11 installations.
On an X11 device, plotting sometimes, e.g., when running
demo("image")
, results in “Error: color allocation error”.
This is an X problem, and only indirectly related to R. It occurs when
applications started prior to R have used all the available colors.
(How many colors are available depends on the X configuration; sometimes
only 256 colors can be used.)
You could also set the
colortype
of
X11()
to
"pseudo.cube"
rather than the default
"pseudo"
. See the
help page for
X11()
for more information.
7.10 How do I convert factors to numeric?
It may happen that when reading numeric data into R (usually, when
reading in a file), they come in as factors. If
is such a
factor object, you can use
as.numeric(as.character(f))
to get the numbers back. More efficient, but harder to remember, is
as.numeric(levels(f))[as.integer(f)]
In any case, do not call
as.numeric()
or their likes directly for
the task at hand (as
as.numeric()
or
unclass()
give the
internal codes).
7.11 Are Trellis displays implemented in R?
The recommended package
lattice
(which is based on base
package
grid
) provides graphical functionality that is compatible
with most Trellis commands.
You could also look at
coplot()
and
dotchart()
which might
do at least some of what you want. Note also that the R version of
pairs()
is fairly general and provides most of the functionality
of
splom()
, and that R’s default plot method has an argument
asp
allowing to specify (and fix against device resizing) the
aspect ratio of the plot.
(Because the word “Trellis” has been claimed as a trademark we do not
use it in R. The name “lattice” has been chosen for the R
equivalent.)
7.12 What are the enclosing and parent environments?
Inside a function you may want to access variables in two additional
environments: the one that the function was defined in (“enclosing”),
and the one it was invoked in (“parent”).
If you create a function at the command line or load it in a package its
enclosing environment is the global workspace. If you define a function
f()
inside another function
g()
its enclosing environment
is the environment inside
g()
. The enclosing environment for a
function is fixed when the function is created. You can find out the
enclosing environment for a function
f()
using
environment(f)
The “parent” environment, on the other hand, is defined when you
invoke a function. If you invoke
lm()
at the command line its
parent environment is the global workspace, if you invoke it inside a
function
f()
then its parent environment is the environment
inside
f()
. You can find out the parent environment for an
invocation of a function by using
parent.frame()
or
sys.frame(sys.parent())
So for most user-visible functions the enclosing environment will be the
global workspace, since that is where most functions are defined. The
parent environment will be wherever the function happens to be called
from. If a function
f()
is defined inside another function
g()
it will probably be used inside
g()
as well, so its
parent environment and enclosing environment will probably be the same.
Parent environments are important because things like model formulas
need to be evaluated in the environment the function was called from,
since that’s where all the variables will be available. This relies on
the parent environment being potentially different with each invocation.
Enclosing environments are important because a function can use
variables in the enclosing environment to share information with other
functions or with other invocations of itself (see the section on
lexical scoping). This relies on the enclosing environment being the
same each time the function is invoked. (In C this would be done with
static variables.)
Scoping
is
hard. Looking at examples helps. It is particularly
instructive to look at examples that work differently in R and S and try
to see why they differ. One way to describe the scoping differences
between R and S is to say that in S the enclosing environment is
always
the global workspace, but in R the enclosing environment
is wherever the function was created.
7.13 How can I substitute into a plot label?
Often, it is desired to use the value of an R object in a plot label,
e.g., a title. This is easily accomplished using
paste()
if the
label is a simple character string, but not always obvious in case the
label is an expression (for refined mathematical annotation). In such a
case, either use
parse()
on your pasted character string or use
substitute()
on an expression. For example, if
ahat
is an
estimator of your parameter
of interest, use
title(substitute(hat(a) == ahat, list(ahat = ahat)))
(note that it is ‘
==
’ and not ‘
’). Sometimes
bquote()
gives a more compact form, e.g.,
title(bquote(hat(a) = .(ahat)))
where subexpressions enclosed in ‘
.()
’ are replaced by their
values.
There are more examples in the mailing list archives.
7.14 What are valid names?
When creating data frames using
data.frame()
or
read.table()
, R by default ensures that the variable names are
syntactically valid. (The argument
check.names
to these
functions controls whether variable names are checked and adjusted by
make.names()
if needed.)
To understand what names are “valid”, one needs to take into account
that the term “name” is used in several different (but related) ways
in the language:
syntactic name
is a string the parser interprets as this type
of expression. It consists of letters, numbers, and the dot and
underscore characters, and starts with either a letter or a dot not
followed by a number. Reserved words are not syntactic names.
An
object name
is a string associated with an object that is
assigned in an expression either by having the object name on the left
of an assignment operation or as an argument to the
assign()
function. It is usually a syntactic name as well, but can be any
non-empty string if it is quoted (and it is always quoted in the call to
assign()
).
An
argument name
is what appears to the left of the equals sign
when supplying an argument in a function call (for example,
f(trim=.5)
). Argument names are also usually syntactic names,
but again can be anything if they are quoted.
An
element name
is a string that identifies a piece of an object
(a component of a list, for example.) When it is used on the right of
the ‘
’ operator, it must be a syntactic name, or quoted.
Otherwise, element names can be any strings. (When an object is used as
a database, as in a call to
eval()
or
attach()
, the
element names become object names.)
Finally, a
file name
is a string identifying a file in the
operating system for reading, writing, etc. It really has nothing much
to do with names in the language, but it is traditional to call these
strings file “names”.
7.15 Are GAMs implemented in R?
Package
gam
from
CRAN
implements all the Generalized
Additive Models (GAM) functionality as described in the GAM chapter of
the White Book. In particular, it implements backfitting with both
local regression and smoothing splines, and is extendable. There is a
gam()
function for GAMs in package
mgcv
, but it is not
an exact clone of what is described in the White Book (no
lo()
for example). Package
gss
can fit spline-based GAMs too. And
if you can accept regression splines you can use
glm()
. For
Gaussian GAMs you can use
bruto()
from package
mda
7.16 Why is the output not printed when I source() a file?
Most R commands do not generate any output. The command
computes the value 2 and returns it; the command
summary(glm(y~x+z, family=binomial))
fits a logistic regression model, computes some summary information and
returns an object of class
"summary.glm"
(see
How should I write summary methods?
).
If you type ‘
1+1
’ or ‘
summary(glm(y~x+z, family=binomial))
’ at
the command line the returned value is automatically printed (unless it
is
invisible()
), but in other circumstances, such as in a
source()
d file or inside a function it isn’t printed unless you
specifically print it.
To print the value use
or
print(summary(glm(y~x+z, family=binomial)))
instead, or use
source(
file
, echo=TRUE)
7.17 Why does outer() behave strangely with my function?
As the help for
outer()
indicates, it does not work on arbitrary
functions the way the
apply()
family does. It requires functions
that are vectorized to work elementwise on arrays. As you can see by
looking at the code,
outer(x, y, FUN)
creates two large vectors
containing every possible combination of elements of
and
and then passes this to
FUN
all at once. Your function
probably cannot handle two large vectors as parameters.
If you have a function that cannot handle two vectors but can handle two
scalars, then you can still use
outer()
but you will need to wrap
your function up first, to simulate vectorized behavior. Suppose your
function is
foo <- function(x, y, happy) {
stopifnot(length(x) == 1, length(y) == 1) # scalars only!
(x + y) * happy
If you define the general function
wrapper <- function(x, y, my.fun, ...) {
sapply(seq_along(x), FUN = function(i) my.fun(x[i], y[i], ...))
then you can use
outer()
by writing, e.g.,
outer(1:4, 1:2, FUN = wrapper, my.fun = foo, happy = 10)
Scalar functions can also be vectorized using
Vectorize()
7.18 Why does the output from
anova()
depend on the order of factors in the model?
In a model such as
~A+B+A:B
, R will report the difference in sums
of squares between the models
~1
~A
~A+B
and
~A+B+A:B
. If the model were
~B+A+A:B
, R would report
differences between
~1
~B
~A+B
, and
~A+B+A:B
. In the first case the sum of squares for
is
comparing
~1
and
~A
, in the second case it is comparing
~B
and
~B+A
. In a non-orthogonal design (i.e., most
unbalanced designs) these comparisons are (conceptually and numerically)
different.
Some packages report instead the sums of squares based on comparing the
full model to the models with each factor removed one at a time (the
famous ‘Type III sums of squares’ from SAS, for example). These do not
depend on the order of factors in the model. The question of which set
of sums of squares is the Right Thing provokes low-level holy wars on
R-help from time to time.
There is no need to be agitated about the particular sums of squares
that R reports. You can compute your favorite sums of squares quite
easily. Any two models can be compared with
anova(
model1
model2
, and
drop1(
model1
will show the sums of
squares resulting from dropping single terms.
7.19 How do I produce PNG graphics in batch mode?
Under a Unix-like, if your installation supports the
type="cairo"
option to the
png()
device there should be no
problems, and the default settings should just work. This option is not
available for versions of R prior to 2.7.0, or without support for
cairo. From R 2.7.0
png()
by default uses the Quartz device
on macOS, and that too works in batch mode.
Earlier versions of the
png()
device used the X11 driver, which
is a problem in batch mode or for remote operation. If you have
Ghostscript you can use
bitmap()
, which produces a PostScript or
PDF file then converts it to any bitmap format supported by Ghostscript.
On some installations this produces ugly output, on others it is
perfectly satisfactory. Many systems now come with Xvfb from
X.Org
(possibly as an optional
install), which is an X11 server that does not require a screen.
7.20 How can I get command line editing to work?
The Unix-like command-line interface to R can only provide the inbuilt
command line editor which allows recall, editing and re-submission of
prior commands provided that the
GNU
readline library is
available at the time R is configured for compilation. Note that the
‘development’ version of readline including the appropriate headers is
needed: users of Linux binary distributions will need to install
packages such as
libreadline-dev
(Debian) or
readline-devel
(Red Hat).
7.21 How can I turn a string into a variable?
If you have
varname <- c("a", "b", "d")
you can do
for
or
assign(varname[1], 2 + 2)
for
or
eval(substitute(lm(y ~ x + variable),
list(variable = as.name(varname[1]))))
for
At least in the first two cases it is often easier to just use a list,
and then you can easily index it by name
vars <- list(a = 1:10, b = rnorm(100), d = LETTERS)
vars[["a"]]
without any of this messing about. This becomes especially true if you
are finding yourself creating and trying to programmatically access
groups of related variables such as
result1
result2
result3
, and so on: instead of fighting against the language to
use
assign(paste0("result", i), process(get(paste0("dataset", i))))
it is much easier to put the related variables in lists and use
result[[i]] <- process(dataset[[i]])
and, eventually,
result <- lapply(dataset, process)
which is easy to replace with
parLapply
for parallel processing.
7.22 Why do lattice/trellis graphics not work?
The most likely reason is that you forgot to tell R to display the
graph. Lattice functions such as
xyplot()
create a graph object,
but do not display it (the same is true of
ggplot2
graphics,
and Trellis graphics in
S-PLUS
). The
print()
method for the
graph object produces the actual display. When you use these functions
interactively at the command line, the result is automatically printed,
but in
source()
or inside your own functions you will need an
explicit
print()
statement.
7.23 How can I sort the rows of a data frame?
To sort the rows within a data frame, with respect to the values in one
or more of the columns, simply use
order()
(e.g.,
DF[order(DF$a, DF[["b"]]), ]
to sort the data frame
DF
on
columns named
and
).
From R 4.4.0,
sort_by()
provides a less verbose alternative
with a formula interface (e.g.,
sort_by(DF, ~a + b)
).
7.24 Why does the help.start() search engine not work?
Since R 2.10.0, the browser-based search engine in
help.start()
is an
HTML
interface to
help.search()
, and should always work.
Before that, the engine utilized a Java applet. In order for this to
function properly, one needed a compatible version of Java installed on
the system and linked to the browser, and both Java
and
JavaScript enabled in the browser.
7.25 Why did my .Rprofile stop working when I updated R?
Did you read the
NEWS
file? For functions that are not in the
base
package you need to specify the correct package namespace,
since the code will be run
before
the packages are loaded. E.g.,
ps.options(horizontal = FALSE)
help.start()
needs to be
grDevices::ps.options(horizontal = FALSE)
utils::help.start()
7.26 Where have all the methods gone?
Many functions, particularly S3 methods, are now hidden in namespaces.
This has the advantage that they cannot be called inadvertently with
arguments of the wrong class, but it makes them harder to view.
To see the code for an S3 method (e.g.,
[.terms
) use
getS3method("[", "terms")
To see the code for an unexported function
foo()
in the namespace
of package
"bar"
use
bar:::foo
. Don’t use these
constructions to call unexported functions in your own code—they are
probably unexported for a reason and may change without warning.
7.27 How can I create rotated axis labels?
To rotate axis labels (using base graphics), you need to use
text()
, rather than
mtext()
, as the latter does not
support
par("srt")
##
Increase bottom margin to make room for rotated labels
par(mar = c(7, 4, 4, 2) + 0.1)
##
Create plot with no x axis and no x axis label
plot(1 : 8, xaxt = "n", xlab = "")
##
Set up x axis with tick marks alone
axis(1, labels = FALSE)
##
Create some text labels
labels <- paste("Label", 1:8, sep = " ")
##
Plot x axis labels at default tick marks
text(1:8, par("usr")[3] - 0.25, srt = 45, adj = 1,
labels = labels, xpd = TRUE)
##
Plot x axis label at line 6 (of 7)
mtext(1, text = "X Axis Label", line = 6)
When plotting the x axis labels, we use
srt = 45
for text
rotation angle,
adj = 1
to place the right end of text at the
tick marks, and
xpd = TRUE
to allow for text outside the plot
region. You can adjust the value of the
0.25
offset as required
to move the axis labels up or down relative to the x axis. See
?par
for more information.
Also see Figure 1 and associated code in
Paul Murrell (2003),
“Integrating grid Graphics Output with Base Graphics Output”,
R News
3/2
, 7–12.
7.28 Why is read.table() so inefficient?
By default,
read.table()
needs to read in everything as character
data, and then try to figure out which variables to convert to numerics
or factors. For a large data set, this takes considerable amounts of
time and memory. Performance can substantially be improved by using the
colClasses
argument to specify the classes to be assumed for the
columns of the table.
7.29 What is the difference between package and library?
package
is a standardized collection of material extending R,
e.g. providing code, data, or documentation. A
library
is a
place (directory) where R knows to find packages it can use (i.e., which
were
installed
). R is told to use a package (to “load” it and
add it to the search path) via calls to the function
library
I.e.,
library()
is employed to load a package from libraries
containing packages.
See
R Add-On Packages
, for more details. See also Uwe Ligges (2003),
“R Help Desk: Package Management”,
R News
3/3
37–39.
7.31 Why doesn’t R think these numbers are equal?
The only numbers that can be represented exactly in R’s numeric type are
integers and fractions whose denominator is a power of 2. All other
numbers are internally rounded to (typically) 53 binary digits accuracy.
As a result, two floating point numbers will not reliably be equal
unless they have been computed by the same algorithm, and not always
even then. For example
R> a <- sqrt(2)
R> a * a == 2
[1] FALSE
R> a * a - 2
[1] 4.440892e-16
R> print(a * a, digits = 18)
[1] 2.00000000000000044
The function
all.equal()
compares two objects using a numeric
tolerance of
.Machine$double.eps ^ 0.5
. If you want much greater
accuracy than this you will need to consider error propagation
carefully.
A discussion with many easily followed examples is in Appendix G
“Computational Precision and Floating Point Arithmetic”, pages
753–771 of
Statistical Analysis and Data Display: An Intermediate
Course with Examples in R
, Richard M. Heiberger and Burt Holland
(Springer 2015, second edition).
This appendix is a free download from
For more information, see e.g.
David Goldberg (1991),
“What Every Computer Scientist Should Know About Floating-Point Arithmetic”,
ACM Computing Surveys
23/1
, 5–48,
also available via
Here is another example, this time using addition:
R> .3 + .6 == .9
[1] FALSE
R> .3 + .6 - .9
[1] -1.110223e-16
R> print(matrix(c(.3, .6, .9, .3 + .6)), digits = 18)
[,1]
[1,] 0.299999999999999989
[2,] 0.599999999999999978
[3,] 0.900000000000000022
[4,] 0.899999999999999911
7.32 How can I capture or ignore errors in a long simulation?
Use
try()
, which returns an object of class
"try-error"
instead of an error, or preferably
tryCatch()
, where the return
value can be configured more flexibly. For example
beta[i,] <- tryCatch(coef(lm(formula, data)),
error = function(e) rep(NaN, 4))
would return the coefficients if the
lm()
call succeeded and
would return
c(NaN, NaN, NaN, NaN)
if it failed (presumably there
are supposed to be 4 coefficients in this example).
7.33 Why are powers of negative numbers wrong?
You are probably seeing something like
and misunderstanding the precedence rules for expressions in R.
Write
to get the square of
-2
The precedence rules are documented in
?Syntax
, and to see how R
interprets an expression you can look at the parse tree
R> as.list(quote(-2^2))
[[1]]
`-`
[[2]]
2^2
7.34 How can I save the result of each iteration in a loop into a separate file?
One way is to use
paste()
(or
sprintf()
) to concatenate a
stem filename and the iteration number while
file.path()
constructs the path. For example, to save results into files
result1.rda
, …,
result100.rda
in the subdirectory
Results
of the current working directory, one can use
for(i in 1:100) {
## Calculations constructing "some_object" ...
fp <- file.path("Results", paste0("result", i, ".rda"))
save(list = "some_object", file = fp)
7.36 Why are there unwanted borders, lines or grid-like artifacts when viewing a plot saved to a PS or PDF file?
This can occur when using functions such as
polygon()
filled.contour()
image()
or other functions which may
call these internally. In the case of
polygon()
, you may observe
unwanted borders between the polygons even when setting the
border
argument to
NA
or
"transparent"
The source of the problem is the PS/PDF viewer when the plot is
anti-aliased. The details for the solution will be different depending
upon the viewer used, the operating system and may change over time.
For some common viewers, consider the following:
Acrobat Reader (cross platform)
There are options in Preferences to enable/disable text smoothing, image
smoothing and line art smoothing.
Disable line art smoothing.
Preview (macOS)
There is an option in Preferences to enable/disable anti-aliasing of
text and line art.
Disable this option.
GSview (cross platform)
There are settings for Text Alpha and Graphics Alpha.
Change Graphics Alpha from 4 bits to 1 bit to disable graphic
anti-aliasing.
gv (Unix-like X)
There is an option to enable/disable anti-aliasing.
Disable this option.
Evince (Linux/GNOME)
There is not an option to disable anti-aliasing in this viewer.
Okular (Linux/KDE)
There is not an option in the GUI to enable/disable anti-aliasing.
From a console command line, use:
$ kwriteconfig --file okularpartrc --group 'Dlg Performance' \
--key GraphicsAntialias Disabled
Then restart Okular. Change the final word to ‘
Enabled
’ to restore
the original setting.
7.37 Why does backslash behave strangely inside strings?
This question most often comes up in relation to file names (see
How do file names work in Windows?
) but it also happens that
people complain that they cannot seem to put a single ‘
’ character
into a text string unless it happens to be followed by certain other
characters.
To understand this, you have to distinguish between character strings
and
representations
of character strings. Mostly, the
representation in R is just the string with a single or double quote at
either end, but there are strings that cannot be represented that way,
e.g., strings that themselves contain the quote character. So
> str <- "This \"text\" is quoted"
> str
[1] "This \"text\" is quoted"
> cat(str, "\n")
This "text" is quoted
The
escape sequences
\"
’ and ‘
\n
’ represent a double
quote and the newline character respectively. Printing text strings,
using
print()
or by typing the name at the prompt will use the
escape sequences too, but the
cat()
function will display the
string as-is. Notice that ‘
"\n"
’ is a one-character string, not
two; the backslash is not actually in the string, it is just generated
in the printed representation.
> nchar("\n")
[1] 1
> substring("\n", 1, 1)
[1] "\n"
So how do you put a backslash in a string? For this, you have to
escape the escape character. I.e., you have to double the backslash.
as in
Some functions, particularly those involving regular expression
matching, themselves use metacharacters, which may need to be escaped by
the backslash mechanism. In those cases you may need a
quadruple
backslash to represent a single literal one.
In versions of R up to 2.4.1 an unknown escape sequence like ‘
\p
was quietly interpreted as just ‘
’. Current versions of R emit a
warning.
7.38 How can I put error bars or confidence bands on my plot?
Some functions will display a particular kind of plot with error bars,
such as the
bar.err()
function in the
agricolae
package, the
plotCI()
function in the
gplots
package,
the
plotCI()
and
brkdn.plot()
functions in the
plotrix
package and the
error.bars()
error.crosses()
and
error.bars.by()
functions in the
psych
package. Within these types of functions, some will
accept the measures of dispersion (e.g.,
plotCI
), some will
calculate the dispersion measures from the raw values (
bar.err
brkdn.plot
), and some will do both (
error.bars
). Still
other functions will just display error bars, like the dispersion
function in the
plotrix
package. Most of the above functions
use the
arrows()
function in the base
graphics
package to
draw the error bars.
The above functions all use the base graphics system. The grid and
lattice graphics systems also have specific functions for displaying
error bars, e.g., the
grid.arrow()
function in the
grid
package, and the
geom_errorbar()
geom_errorbarh()
geom_pointrange()
geom_linerange()
geom_crossbar()
and
geom_ribbon()
functions in the
ggplot2
package. In the lattice system, error bars can be
displayed with
Dotplot()
or
xYplot()
in the
Hmisc
package and
segplot()
in the
latticeExtra
package.
7.39 How do I create a plot with two y-axes?
Creating a graph with two y-axes, i.e., with two sorts of data that are
scaled to the same vertical size and showing separate vertical axes on
the left and right sides of the plot that reflect the original scales of
the data, is possible in R but is not recommended. The basic approach
for constructing such graphs is to use
par(new=TRUE)
(see
?par
); functions
twoord.plot()
(in the
plotrix
package) and
doubleYScale()
(in the
latticeExtra
package) automate the process somewhat.
7.40 How do I access the source code for a function?
In most cases, typing the name of the function will print its source
code. However, code is sometimes hidden in a namespace, or compiled. For
a complete overview on how to access source code, see Uwe Ligges (2006),
“Help Desk: Accessing the sources”,
R News
6/4
43–45 (
).
7.41 Why does summary() report strange results for the R^2 estimate when I fit a linear model with no intercept?
As described in
?summary.lm
, when the intercept is zero (e.g.,
from
y ~ x - 1
or
y ~ x + 0
),
summary.lm()
uses the
formula
R^2 = 1 - Sum(R[i]^2) / Sum((y[i])^2),
which is different from the usual
R^2 = 1 - Sum(R[i]^2) / Sum((y[i] - mean(y))^2).
There are several reasons for this:
Otherwise the
R^2
could be negative (because the model with zero
intercept can fit
worse
than the constant-mean model it is
implicitly compared to).
If you set the slope to zero in the model with a line through the
origin you get fitted values y*=0
The model with constant, non-zero mean is not nested in the model
with a line through the origin.
All these come down to saying that if you know
a priori
that
E[Y]=0
when
x=0
then the ‘null’ model that you should
compare to the fitted line, the model where
doesn’t explain any
of the variance, is the model where
E[Y]=0
everywhere. (If you
don’t know a priori that
E[Y]=0
when
x=0
, then you
probably shouldn’t be fitting a line through the origin.)
7.42 Why is R apparently not releasing memory?
This question is often asked in different flavors along the lines of
“I have removed objects in R and run
gc()
and yet
ps
top
still shows the R process using a lot of
memory”, often on Linux machines.
This is an artifact of the way the operating system (OS) allocates
memory. In general it is common that the OS is not capable of
releasing all unused memory. In extreme cases it is possible that even
if R frees almost all its memory, the OS can not release any of it due
to its design and thus tools such as
ps
or
top
will
report substantial amount of resident RAM used by the R process even
though R has released all that memory. In general such tools do
not
report the actual memory usage of the process but rather
what the OS is reserving for that process.
The short answer is that this is a limitation of the memory allocator
in the operating system and there is nothing R can do about it. That
space is simply kept by the OS in the hope that R will ask for it
later. The following paragraph gives more in-depth answer with
technical details on how this happens.
Most systems use two separate ways to allocate memory. For allocation
of large chunks they will use
mmap
to map memory into the
process address space. Such chunks can be released immediately when
they are completely free, because they can reside anywhere in the
virtual memory. However, this is a relatively expensive operation and
many OSes have a limit on the number of such allocated chunks, so this
is only used for allocating large memory regions. For smaller
allocations the system can expand the data segment of the process
(historically using the
brk
system call), but this whole area
is always contiguous. The OS can only move the end of this space, it
cannot create any “holes”. Since this operation is fairly cheap, it
is used for allocations of small pieces of memory. However, the
side-effect is that even if there is just one byte that is in use
at the end of the data segment, the OS cannot release any memory
at all, because it cannot change the address of that byte. This is
actually more common than it may seem, because allocating a lot of
intermediate objects, then allocating a result object and removing all
intermediate objects is a very common practice. Since the result is
allocated at the end it will prevent the OS from releasing any memory
used by the intermediate objects. In practice, this is not necessarily
a problem, because modern operating systems can page out unused
portions of the virtual memory so it does not necessarily reduce the
amount of real memory available for other applications. Typically,
small objects such as strings or pairlists will be affected by this
behavior, whereas large objects such as long vectors will be allocated
using
mmap
and thus not affected. On Linux (and possibly other
Unix-like systems) it is possible to use the
mallinfo
system call
(also see the
mallinfo
package) to
query the allocator about the layout of the allocations, including the
actually used memory as well as unused memory that cannot be released.
7.43 How can I enable secure https downloads in R?
From R 4.2.0,
"libcurl"
download method is always available and
used for
HTTPS
by default on all platforms. It has been used
since R 3.3.0 everywhere but Windows where the default method
"wininet"
also supported
HTTPS
So nothing needs to be done to access ‘
’ websites in recent
versions of R.
7.44 How can I get CRAN package binaries for outdated versions of R?
Since March 2016, Windows and macOS binaries of
CRAN
packages for old
versions of R (released more than 5 years ago) are made available from a
central
CRAN
archive server instead of the
CRAN
mirrors. To get
these, one should set the
CRAN
“mirror” element of the
repos
option accordingly, by something like
local({r <- getOption("repos")
r["CRAN"] <- "http://CRAN-archive.R-project.org"
options(repos = r)
})
(see
?options
for more information).
9 R Bugs
9.1 What is a bug?
If R executes an illegal instruction, or dies with an operating system
error message that indicates a problem in the program (as opposed to
something like “disk full”), then it is certainly a bug. If you call
.C()
.Fortran()
.External()
or
.Call()
(or
.Internal()
) yourself (or in a function you wrote), you can
always crash R by using wrong argument types (modes). This is not a
bug.
Taking forever to complete a command can be a bug, but you must make
certain that it was really R’s fault. Some commands simply take a long
time. If the input was such that you
know
it should have been
processed quickly, report a bug. If you don’t know whether the command
should take a long time, find out by looking in the manual or by asking
for assistance.
If a command you are familiar with causes an R error message in a case
where its usual definition ought to be reasonable, it is probably a bug.
If a command does the wrong thing, that is a bug. But be sure you know
for certain what it ought to have done. If you aren’t familiar with the
command, or don’t know for certain how the command is supposed to work,
then it might actually be working right. For example, people sometimes
think there is a bug in R’s mathematics because they don’t understand
how finite-precision arithmetic works. Rather than jumping to
conclusions, show the problem to someone who knows for certain.
Unexpected results of comparison of decimal numbers, for example
0.28 * 100 != 28
or
0.1 + 0.2 != 0.3
, are not a bug.
See
Why doesn’t R think these numbers are equal?
, for more details.
Finally, a command’s intended definition may not be best for statistical
analysis. This is a very important sort of problem, but it is also a
matter of judgment. Also, it is easy to come to such a conclusion out
of ignorance of some of the existing features. It is probably best not
to complain about such a problem until you have checked the
documentation in the usual ways, feel confident that you understand it,
and know for certain that what you want is not available. If you are
not sure what the command is supposed to do after a careful reading of
the manual this indicates a bug in the manual. The manual’s job is to
make everything clear. It is just as important to report documentation
bugs as program bugs. However, we know that the introductory
documentation is seriously inadequate, so you don’t need to report this.
If the online argument list of a function disagrees with the manual, one
of them must be wrong, so report the bug.
See also “Making sure it’s a bug” in
Bug Reporting in R
for more
information.
9.2 How to report a bug
When you decide that there is a bug, it is important to report it and to
report it in a way which is useful. What is most useful is an exact
description of what commands you type, starting with the shell command
to run R, until the problem happens. Always include the version of R,
machine, and operating system that you are using; type
version
in
R to print this.
The most important principle in reporting a bug is to report
facts
, not hypotheses or categorizations. It is always easier to
report the facts, but people seem to prefer to strain to posit
explanations and report them instead. If the explanations are based on
guesses about how R is implemented, they will be useless; others will
have to try to figure out what the facts must have been to lead to such
speculations. Sometimes this is impossible. But in any case, it is
unnecessary work for the ones trying to fix the problem.
For example, suppose that on a data set which you know to be quite large
the command
R> data.frame(x, y, z, monday, tuesday)
never returns. Do not report that
data.frame()
fails for large
data sets. Perhaps it fails when a variable name is a day of the week.
If this is so then when others got your report they would try out the
data.frame()
command on a large data set, probably with no day of
the week variable name, and not see any problem. There is no way in the
world that others could guess that they should try a day of the week
variable name.
Or perhaps the command fails because the last command you used was a
method for
"["()
that had a bug causing R’s internal data
structures to be corrupted and making the
data.frame()
command
fail from then on. This is why others need to know what other commands
you have typed (or read from your startup file).
It is very useful to try and find simple examples that produce
apparently the same bug, and somewhat useful to find simple examples
that might be expected to produce the bug but actually do not. If you
want to debug the problem and find exactly what caused it, that is
wonderful. You should still report the facts as well as any
explanations or solutions. Please include an example that reproduces
(e.g.,
) the problem,
preferably the simplest one you have found.
Invoking R with the
--vanilla
option may help in isolating a
bug. This ensures that the site profile and saved data files are not
read.
Before you actually submit a bug report, you should check whether the
bug has already been reported and/or fixed. First, try the “Show open
bugs new-to-old” or the search facility on
. Second, consult
, which
records changes that will appear in the
next
release of R,
including bug fixes that do not appear on the Bug Tracker.
Third, if possible try the current r-patched or r-devel version of R.
If a bug has already been reported or fixed, please do not submit
further bug reports on it. Finally, check carefully whether the bug is
with R, or a contributed package. Bug reports on contributed packages
should be sent first to the package maintainer, and only submitted to
the R-bugs repository by package maintainers, mentioning the package in
the subject line.
A bug report can be generated using the function
bug.report()
For reports on R this will open the R Bugzilla page at
: for a contributed package it will open
the package’s bug tracker Web page or help you compose an email to the
maintainer. Since 2016, only “members” (including all who have
previously submitted bugs) can submit new bugs on the R Bugzilla. See
“Where to submit bug reports and patches” on
Bug Reporting in R
for more
information.
There is a section of the bug repository for suggestions for
enhancements for R labelled ‘
wishlist
’. Suggestions can be
submitted in the same ways as bugs, but please ensure that the subject
line makes clear that this is for the wishlist and not a bug report, for
example by starting with ‘
Wishlist:
’.
Comments on and suggestions for the Windows port of R should be sent to
R-windows@R-project.org
Corrections to and comments on message translations should be sent to the
last translator (listed at the top of the appropriate ‘
.po
’ file)
or to the translation team as listed at