|
 Initially called Passtec+, the project of converting Visual Basic algorithms implemented in Passtec into a library of function for Splus and R is a PNEC-Art4/IFREMER initiative (Benoit Beliaeff and Frédéric Ibanez). It was ultimately renamed Pastecs to avoid conflicting name with other software. The Pastecs library is entirely written in S language by Philippe Grosjean. It contains a hundred functions to regulate, decompose and analyze uni- or multivariate space-time series. An intuitive user interface with menus and dialog boxes is currently in development. |
|||||||||||
|
||||||||||||
 |
for Splus 2000 release 3 (library and installation instructions) |
|
for other Splus versions (source code) |
|
for R 2.0.0 or above, Windows (download and installation instructions) |
|
for R 1.7.2 or above, Macintosh 8.x-9.1 (idem) |
|
for R 2.0.0 or above, Unix/Linux/MacOS X (source code and installation instructions) |
Download and install Pastecs for Splus 2000 release 3
To use this version of Pastecs, you need Splus 2000 professional for Windows (the standard version does not provide a command window, and thus, does not allow full access to the function libraries). Please, verify your Splus version in the help -> about... box, and upgrade to release 3 if it is an earlier release. Look at Insightful web site for getting required files.
Run the Pastecs installation file to place the library in your <Splus dir>\library subdirectory (usually c:\program files\sp2000\library). Once it is done, you can access Pastecs functions from within the command window of Splus by typing:
> library(pastecs)Online help is available using the function help() or ?. For instance, to get help on the tsd() function, enter:
> ?tsdFrom there, you can copy and paste examples to the command window to run them. An index of all functions available in the Pastecs library is available in an ASCII file located in <Splus dir>\library\pastecs\index.txt, once the software is installed. For more information on how to use Pastecs, read the user manual (in French only).
Pastecs Splus installation, v. 1.0-1 (.exe, 455 Ko).
Pastecs user manual in French (.pdf, 3556 Ko).
Examples of the user manual for Splus (.ssc, 20 Ko).
Last update: 12/11/2002.
Download Pastecs code
source for other Splus versions
Pastecs was not tested on other versions of Splus than version 2000 release 3 and the library is not compiled for Unix/Linux versions of Splus. However, we provide the source code (functions, data files, online help and companion files). For these versions, you should compile the library yourself. Read the Splus manual of your particular version to know how to compile the library on your system.
Once you have compiled and installed the library, read the instructions on using Pastecs on Splus 2000 (hereabove), that should also apply to your particular Splus version. A user manual (in French only) is also provided to help you starting to use the Pastecs functions. Please, report your comments on using Pastecs on other Splus versions to Philippe Grosjean. Pastecs is now totally unsupported on other environments than R!
Pastecs Splus code source, v. 1.0-1 (.zip, 269 Ko).
Pastecs user manual in French (.pdf, 3556 Ko).
Examples of the user manual for Splus (.ssc, 20 Ko).
Last update: 12/11/2002.
Download and install Pastecs
for R 2.0.0 or above on Windows
First of all, get the latest R binaries for Windows from http://cran-r.project.org and install them on your computer (it is free!). Once R is running on your system, you have the choice to install the Pastecs library from CRAN or from the SciViews site. To install Pastecs from CRAN, connect to the internet from the machine where you installed R, start R (Rgui.exe), click in its menu Packages -> Install package from CRAN... and select pastecs in the list. That's it! Note that, once installed, you can update Pastecs as well as other packages easily from within R by clicking on Packages -> Update packages from CRAN... Do it regularly to make sure you are always working with the most up-to-date packages!
To install Pastecs using the version provided in this site, or to install/update the Pastecs package on a computer that is not connected to internet, download the Pastecs library in .zip format hereunder and copy it to a temporary directory on the computer where you installed R. Start R (Rgui.exe) and click Packages -> Install package from local zip file..., select the file and click OK. When the package is installed, you can delete the .zip file.
Once the Pastecs library is installed, you can access its functions in a session after entering:
> library(pastecs)Online help is available using the R function help() or ?. For instance, to get help on the tsd() function, enter:
> ?tsdTo run examples for the same function, type:
> example(tsd)and finally, to browse the online help in html format, go to Help -> R language (html), menu of Rgui.exe. In the help main page, select Packages and then pastecs. You can learn more about how to use Pastecs in the user manual (in French only).
Pastecs R library for Windows, v. 1.2-0 (.zip, 1.4 Mb, including the user manual).
Examples of the user manual for R (.R, 20 Ko).
Last update: 10/01/2005.
Download and install Pastecs for R 1.7.2 or above on Macintosh 8.x-9.1
If you use a Macintosh with an old operating system 8.x-9.1, then you can only install an old, obsolete version of R (1.7.2). R is now only maintained for MacOS X.
To install that obsolete native version of R on your Macintosh, download and unstuff the installation file (rm172.sit) on your fastest hard disk. You need the full version (base + recommended packages) to run all Pastecs functions. Once it is done, download the Pastecs R library for Macintosh, hereunder, and place it in the library subdirectory of your R installation. Unstuff it there. Once it is done, you can delete all .sit installation files.
To verify that both R and the Pastecs library are correctly installed, start R and load the Pastecs library by entering:
> library(pastecs)(note that this command must be reentered on each session). If the command issues without errors, the Pastecs library should be correctly installed, otherwise, verify that you unstuffed the Pastecs installation file in the correct library subdirectory of your R installation. You still have to update html help to include Pastecs help files. Enter:
> link.html.help()Pastecs is now completely installed. Online help is available from the help menu, or by using the function help() or ?. For instance, to get help on the tsd() function, enter:
> ?tsdTo run examples for the same function, type:
> example(tsd)You can learn more about how to use Pastecs in the user manual (in French only).
Pastecs R library for Macintosh, v. 1.1-1 (.sit, 306 Ko).
Pastecs user manual in French (.pdf, 3556 Ko).
Examples of the user manual for R (.R, 20 Ko).
Last update: 10/01/2005.
Compile and install Pastecs for R 2.0.0 or above on Unix/Linux/MacOS X
For Unix/Linux/MacOS X boxes, the Pastecs library is provided as source code only. You have to compile it in your own particular system. This is not difficult. First, install the latest R version for your system. You find several precompiled binaires of R for various Unix/Linux systems on CRAN (http://cran.r-project.org). If there is no binaries for your particular system, you still can compile it from sources. Browse the CRAN web site for finding how to do this.
Once R is installed, download the .tar.gz source file of the Pastecs library, either from CRAN, or from hereunder in a directory, let's say /usr/temp/. Compilation and installation is done by a single command:
R CMD INSTALL /usr/temp/pastecs_1.2-0.tar.gzThat's all you have to do! Online help is available from your R console by using the function help() or ?. For instance, to get help on the tsd() function, enter:
> ?tsdTo run examples for the same function, type:
> example(tsd)You can learn more about how to use Pastecs in the user manual (in French only).
Pastecs R library source code, v. 1.2-0 (.tar.gz, 1.4 Mo, including the user manual).
Examples of the user manual for R (.R, 20 Ko).
Last update: 10/01/2005.
| [ HOME ] [ SciViews-R ] [ Tinn-R] [ Pastecs ] [ ShellAxis ] [ LaboKit ] [ Benchmark] [ Links ] |