Tag: graphics

Top Ten Graphs for Business Analytics -Pie Charts (1/10)

I have not been really posting or writing worthwhile on the website for some time, as I am still busy writing ” R for Business Analytics” which I hope to get out before year end. However while doing research for that, I came across many types of graphs and what struck me is the actual usage of some kinds of graphs is very different in business analytics as compared to statistical computing.

The criterion of top ten graphs is as follows-

1) Usage-The order in which they appear is not strictly in terms of desirability but actual frequency of usage. So a frequently used graph like box plot would be recommended above say a violin plot.

2) Adequacy- Data Visualization paradigms change over time- but the need for accurate conveying of maximum information in a minium space without overwhelming reader or misleading data perceptions.

3) Ease of creation- A simpler graph created by a single function is more preferrable to writing 4-5 lines of code to create an elaborate graph.

4) Aesthetics– Aesthetics is relative and  in addition studies have shown visual perception varies across cultures and geographies. However , beauty is universally appreciated and a pretty graph is sometimes and often preferred over a not so pretty graph. Here being pretty is in both visual appeal without compromising perceptual inference from graphical analysis.

 

so When do we use a bar chart versus a line graph versus a pie chart? When is a mosaic plot more handy and when should histograms be used with density plots? The list tries to capture most of these practicalities.

Let me elaborate on some specific graphs-

1) Pie Chart- While Pie Chart is not really used much in stats computing, and indeed it is considered a misleading example of data visualization especially the skewed or two dimensional charts. However when it comes to evaluating market share at a particular instance, a pie chart is simple to understand. At the most two pie charts are needed for comparing two different snapshots, but three or more pie charts on same data at different points of time is definitely a bad case.

In R you can create piechart, by just using pie(dataset$variable)

As per official documentation, pie charts are not  recommended at all.

http://stat.ethz.ch/R-manual/R-patched/library/graphics/html/pie.html

Pie charts are a very bad way of displaying information. The eye is good at judging linear measures and bad at judging relative areas. A bar chart or dot chart is a preferable way of displaying this type of data.

Cleveland (1985), page 264: “Data that can be shown by pie charts always can be shown by a dot chart. This means that judgements of position along a common scale can be made instead of the less accurate angle judgements.” This statement is based on the empirical investigations of Cleveland and McGill as well as investigations by perceptual psychologists.

—-

Despite this, pie charts are frequently used as an important metric they inevitably convey is market share. Market share remains an important analytical metric for business.

The pie3D( ) function in the plotrix package provides 3D exploded pie charts.An exploded pie chart remains a very commonly used (or misused) chart.

From http://lilt.ilstu.edu/jpda/charts/chart%20tips/Chartstip%202.htm#Rules

we see some rules for using Pie charts.

 

  1. Avoid using pie charts.
  2. Use pie charts only for data that add up to some meaningful total.
  3. Never ever use three-dimensional pie charts; they are even worse than two-dimensional pies.
  4. Avoid forcing comparisons across more than one pie chart

 

From the R Graph Gallery (a slightly outdated but still very comprehensive graphical repository)

http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=4

par(bg="gray")
pie(rep(1,24), col=rainbow(24), radius=0.9)
title(main="Color Wheel", cex.main=1.4, font.main=3)
title(xlab="(test)", cex.lab=0.8, font.lab=3)
(Note adding a grey background is quite easy in the basic graphics device as well without using an advanced graphical package)

 

Related Articles

Using Views in R and comparing functions across multiple packages

Some RDF hacking relating to updating probabil...
Image via Wikipedia

R has almost 2923 available packages

This makes the task of searching among these packages and comparing functions for the same analytical task across different packages a bit tedious and prone to manual searching (of reading multiple Pdfs of help /vignette of packages) or sending an email to the R help list.

However using R Views is a slightly better way of managing all your analytical requirements for software rather than the large number of packages (see Graphics view below).

CRAN Task Views allow you to browse packages by topic and provide tools to automatically install all packages for special areas of interest. Currently, 28 views are available. http://cran.r-project.org/web/views/

Bayesian Bayesian Inference
ChemPhys Chemometrics and Computational Physics
ClinicalTrials Clinical Trial Design, Monitoring, and Analysis
Cluster Cluster Analysis & Finite Mixture Models
Distributions Probability Distributions
Econometrics Computational Econometrics
Environmetrics Analysis of Ecological and Environmental Data
ExperimentalDesign Design of Experiments (DoE) & Analysis of Experimental Data
Finance Empirical Finance
Genetics Statistical Genetics
Graphics Graphic Displays & Dynamic Graphics & Graphic Devices & Visualization
gR gRaphical Models in R
HighPerformanceComputing High-Performance and Parallel Computing with R
MachineLearning Machine Learning & Statistical Learning
MedicalImaging Medical Image Analysis
Multivariate Multivariate Statistics
NaturalLanguageProcessing Natural Language Processing
OfficialStatistics Official Statistics & Survey Methodology
Optimization Optimization and Mathematical Programming
Pharmacokinetics Analysis of Pharmacokinetic Data
Phylogenetics Phylogenetics, Especially Comparative Methods
Psychometrics Psychometric Models and Methods
ReproducibleResearch Reproducible Research
Robust Robust Statistical Methods
SocialSciences Statistics for the Social Sciences
Spatial Analysis of Spatial Data
Survival Survival Analysis
TimeSeries Time Series Analysis

To automatically install these views, the ctv package needs to be installed, e.g., via

install.packages("ctv")
library("ctv")
Created by Pretty R at inside-R.org


and then the views can be installed via install.views or update.views (which first assesses which of the packages are already installed and up-to-date), e.g.,

install.views("Econometrics")
 update.views("Econometrics")
 Created by Pretty R at inside-R.org

CRAN Task View: Graphic Displays & Dynamic Graphics & Graphic Devices & Visualization

Maintainer: Nicholas Lewin-Koh
Contact: nikko at hailmail.net
Version: 2009-10-28

R is rich with facilities for creating and developing interesting graphics. Base R contains functionality for many plot types including coplots, mosaic plots, biplots, and the list goes on. There are devices such as postscript, png, jpeg and pdf for outputting graphics as well as device drivers for all platforms running R. lattice and grid are supplied with R’s recommended packages and are included in every binary distribution. lattice is an R implementation of William Cleveland’s trellis graphics, while grid defines a much more flexible graphics environment than the base R graphics.

R’s base graphics are implemented in the same way as in the S3 system developed by Becker, Chambers, and Wilks. There is a static device, which is treated as a static canvas and objects are drawn on the device through R plotting commands. The device has a set of global parameters such as margins and layouts which can be manipulated by the user using par() commands. The R graphics engine does not maintain a user visible graphics list, and there is no system of double buffering, so objects cannot be easily edited without redrawing a whole plot. This situation may change in R 2.7.x, where developers are working on double buffering for R devices. Even so, the base R graphics can produce many plots with extremely fine graphics in many specialized instances.

One can quickly run into trouble with R’s base graphic system if one wants to design complex layouts where scaling is maintained properly on resizing, nested graphs are desired or more interactivity is needed. grid was designed by Paul Murrell to overcome some of these limitations and as a result packages like lattice, ggplot2, vcd or hexbin (on Bioconductor ) use grid for the underlying primitives. When using plots designed with grid one needs to keep in mind that grid is based on a system of viewports and graphic objects. To add objects one needs to use grid commands, e.g., grid.polygon() rather than polygon(). Also grid maintains a stack of viewports from the device and one needs to make sure the desired viewport is at the top of the stack. There is a great deal of explanatory documentation included with grid as vignettes.

The graphics packages in R can be organized roughly into the following topics, which range from the more user oriented at the top to the more developer oriented at the bottom. The categories are not mutually exclusive but are for the convenience of presentation:

  • Plotting : Enhancements for specialized plots can be found in plotrix, for polar plotting, vcd for categorical data, hexbin (on Bioconductor ) for hexagon binning, gclus for ordering plots and gplots for some plotting enhancements. Some specialized graphs, like Chernoff faces are implemented in aplpack, which also has a nice implementation of Tukey’s bag plot. For 3D plots lattice, scatterplot3d and misc3d provide a selection of plots for different kinds of 3D plotting. scatterplot3d is based on R’s base graphics system, while misc3d is based on rgl. The package onion for visualizing quaternions and octonions is well suited to display 3D graphics based on derived meshes.
  • Graphic Applications : This area is not much different from the plotting section except that these packages have tools that may not for display, but can aid in creating effective displays. Also included are packages with more esoteric plotting methods. For specific subject areas, like maps, or clustering the excellent task views contributed by other dedicated useRs is an excellent place to start.
    • Effect ordering : The gclus package focuses on the ordering of graphs to accentuate cluster structure or natural ordering in the data. While not for graphics directly cba and seriation have functions for creating 1 dimensional orderings from higher dimensional criteria. For ordering an array of displays, biclust can be useful.
    • Large Data Sets : Large data sets can present very different challenges from moderate and small datasets. Aside from overplotting, rendering 1,000,000 points can tax even modern GPU’s. For univariate datalvplot produces letter value boxplots which alleviate some of the problems that standard boxplots exhibit for large data sets. For bivariate data ash can produce a bivariate smoothed histogram very quickly, and hexbin, on Bioconductor , can bin bivariate data onto a hexagonal lattice, the advantage being that the irregular lines and orientation of hexagons do not create linear artifacts. For multivariate data, hexbin can be used to create a scatterplot matrix, combined with lattice. An alternative is to use scagnostics to produce a scaterplot matrix of “data about the data”, and look for interesting combinations of variables.
    • Trees and Graphs : ape and ade4 have functions for plotting phylogenetic trees, which can be used for plotting dendrograms from clustering procedures. While these packages produce decent graphics, they do not use sophisticated algorithms for node placement, so may not be useful for very large trees. igraph has the Tilford-Rheingold algorithm implementead and is useful for plotting larger trees. diagram as facilities for flow diagrams and simple graphs. For more sophisticated graphs Rgraphviz and igraph have functions for plotting and layout, especially useful for representing large networks.
  • Graphics Systems : lattice is built on top of the grid graphics system and is an R implementation of William Cleveland’s trellis system for S-PLUS. lattice allows for building many types of plots with sophisticated layouts based on conditioning. ggplot2 is an R implementation of the system described in “A Grammar of Graphics” by Leland Wilkinson. Like lattice, ggplot (also built on top of grid) assists in trellis-like graphics, but allows for much more. Since it is built on the idea of a semantics for graphics there is much more emphasis on reshaping data, transformation, and assembling the elements of a plot.
  • Devices : Whereas grid is built on top of the R graphics engine, many in the R community have found the R graphics engine somewhat inflexible and have written separate device drivers that either emphasize interactivity or plotting in various graphics formats. R base supplies devices for PostScript, PDF, JPEG and other formats. Devices on CRAN include cairoDevice which is a device based libcairo, which can actually render to many device types. The cairo device is desgned to work with RGTK2, which is an interface to the Gimp Tool Kit, similar to pyGTK2. GDD provides device drivers for several bitmap formats, including GIF and BMP. RSvgDevice is an SVG device driver and interfaces well with with vector drawing programs, or R web development packages, such as Rpad. When SVG devices are for web display developers should be aware that internet explorer does not support SVG, but has their own standard. Trust Microsoft. rgl provides a device driver based on OpenGL, and is good for 3D and interactive development. Lastly, the Augsburg group supplies a set of packages that includes a Java-based device, JavaGD.
  • Colors : The package colorspace provides a set of functions for transforming between color spaces and mixcolor() for mixing colors within a color space. Based on the HCL colors provided in colorspace, vcdprovides a set of functions for choosing color palettes suitable for coding categorical variables ( rainbow_hcl()) and numerical information ( sequential_hcl(), diverge_hcl()). Similar types of palettes are provided in RColorBrewer and dichromat is focused on palettes for color-impaired viewers.
  • Interactive Graphics : There are several efforts to implement interactive graphics systems that interface well with R. In an interactive system the user can interactively query the graphics on the screen with the mouse, or a moveable brush to zoom, pan and query on the device as well as link with other views of the data. rggobi embeds the GGobi interactive graphics system within R, so that one can display a data frame or several in GGobi directly from R. The package has functions to support longitudinal data, and graphs using GGobi’s edge set functionality. The RoSuDA repository maintained and developed by the University of Augsburg group has two packages, iplots and iwidgets as well as their Java development environment including a Java device, JavaGD. Their interactive graphics tools contain functions for alpha blending, which produces darker shading around areas with more data. This is exceptionally useful for parallel coordinate plots where many lines can quickly obscure patterns. playwith has facilities for building interactive versions of R graphics using the cairoDevice and RGtk2. Lastly, the rgl package has mechanisms for interactive manipulation of plots, especially 3D rotations and surfaces.
  • Development : For development of specialized graphics packages in R, grid should probably be the first consideration for any new plot type. rgl has better tools for 3D graphics, since the device is interactive, though it can be slow. An alternative is to use Java and the Java device in the RoSuDA packages, though Java has its own drawbacks. For porting plotting code to grid, using the package gridBase presents a nice intermediate step to embed base graphics in grid graphics and vice versa.

 

Related Articles

Top 10 Games on Linux -sudo update

The phrase "Doom clone" was initiall...
Image via Wikipedia

Here are some cool games I like to play on my Ubuntu 10.10 – I think they run on most other versions of Linux as well. 1) Open Arena– First person Shooter– This is like Quake Arena- very very nice graphics and good for playing for a couple of hours while taking a break. It is available here- http://openarena.ws/smfnews.php ideally if you have a bunch of gaming friends, playing on a local network or internet is quite mind blowing entertaining. And it’s free! 2) Armagetron– This is based on the TRON game of light cycles-It is available here at http://www.armagetronad.net/ or you can use Synaptic packages manager for all the games mentioned here

3) Sudoko–

If violence or cars is not your stuff and you like puzzles like Sudoko, well just install the application Sudoko from http://gnome-sudoku.sourceforge.net/ Also recommended for people of various ages as it has multiple levels.

4) Pinball

If you ever liked Pinball play the open source version from download at http://pinball.sourceforge.net/ Alternatively you can go to Ubuntu Software Centre>Games>Arcade>Emilio>Pinball and you can also build your own pinball if you like the game well enough. 5) Pacman/Njam- Clone of the original classic game.  Downloadable from http://www.linuxcompatible.org/news/story/pacman_for_linux.html 6) Gweled– This is free clone version of Bejeweled. It now has a new website at http://gweled.org/ http://linux.softpedia.com/progDownload/Gweled-Download-3449.html

Gweled is a GNOME version of a popular PalmOS/Windows/Java game called “Bejeweled” or “Diamond Mine”. The aim of the game is to make alignment of 3 or more gems, both vertically or horizontally by swapping adjacent gems. The game ends when there are no possible moves left. Here are some key features of “Gweled”: · exact same gameplay as the commercial versions · SVG original graphics

7) Hearts – For this card game classis you can use Ubuntu software to install the package or go to http://linuxappfinder.com/package/gnome-hearts 8) Card Games- KPatience has almost 14 card games including solitaire, and free cell. 9) Sauerbraten -First person shooter with good network play, edit maps capabilities. You can read more here- http://sauerbraten.org/ 10) Tetris-KBlocks Tetris is the classic game. If you like classic slow games- Tetris is the best. and I like the toughest Tetris game -Bastet http://fph.altervista.org/prog/bastet.html Even an xkcd toon for it

That’s all for holiday season folks, the top 10 lists is based on almost 3 decades of gaming experience, but beauty is the eye of the beholder- so happy gaming for free.
Related Articles

Whats new in the latest version of R

Figure 2: The argument φ and modulus r locate ...
Image via Wikipedia
CHANGES IN R VERSION 2.12.2: http://cran.r-project.org/src/base/NEWS

  SIGNIFICANT USER-VISIBLE CHANGES:

    • Complex arithmetic (notably z^n for complex z and integer n) gave
      incorrect results since R 2.10.0 on platforms without C99 complex
      support.  This and some lesser issues in trignometric functions
      have been corrected.

      Such platforms were rare (we know of Cygwin and FreeBSD).
      However, because of new compiler optimizations in the way complex
      arguments are handled, the same code was selected on x86_64 Linux
      with gcc 4.5.x at the default -O2 optimization (but not at -O).

    • There is a workaround for crashes seen with several packages on
      systems using zlib 1.2.5: see the INSTALLATION section.

  NEW FEATURES:

    • PCRE has been updated to 8.12 (two bug-fix releases since 8.10).

    • rep(), seq(), seq.int() and seq_len() report more often when the
      first element is taken of an argument of incorrect length.

    • The Cocoa back-end for the quartz() graphics device on Mac OS X
      provides a way to disable event loop processing temporarily
      (useful, e.g., for forked instances of R).

    • kernel()'s default for m was not appropriate if coef was a set of
      coefficients.  (Reported by Pierre Chausse.)

    • bug.report() has been updated for the current R bug tracker,
      which does not accept emailed submissions.

    • R CMD check now checks for the correct use of $(LAPACK_LIBS) (as
      well as $(BLAS_LIBS)), since several CRAN recent submissions have
      ignored ‘Writing R Extensions’.

  INSTALLATION:

    • The zlib sources in the distribution are now built with all
      symbols remapped: this is intended to avoid problems seen with
      packages such as XML and rggobi which link to zlib.so.1 on
      systems using zlib 1.2.5.

    • The default for FFLAGS and FCFLAGS with gfortran on x86_64 Linux
      has been changed back to -g -O2: however, setting -g -O may still
      be needed for gfortran 4.3.x.

  PACKAGE INSTALLATION:

    • A LazyDataCompression field in the DESCRIPTION file will be used
      to set the value for the --data-compress option of R CMD INSTALL.

    • Files R/sysdata.rda of more than 1Mb are now stored in the
      lazyload daabase using xz compression: this for example halves
      the installed size of package Imap.

    • R CMD INSTALL now ensures that directories installed from inst
      have search permission for everyone.

      It no longer installs files inst/doc/Rplots.ps and
      inst/doc/Rplots.pdf.  These are almost certainly left-overs from
      Sweave runs, and are often large.

  DEPRECATED & DEFUNCT:

    • The ‘experimental’ alternative specification of a name space via
      .Export() etc is now deprecated.

    • zip.file.extract() is now deprecated.

    • Zip-ing data sets in packages (and hence R CMD INSTALL
      --use-zip-data and the ZipData: yes field in a DESCRIPTION file)
      is deprecated: using efficiently compressed .rda images and
      lazy-loading of data has superseded it.

  BUG FIXES:

    • identical() could in rare cases generate a warning about
      non-pairlist attributes on CHARSXPs.  As these are used for
      internal purposes, the attribute check should be skipped.
      (Reported by Niels Richard Hansen).

    • If the filename extension (usually .Rnw) was not included in a
      call to Sweave(), source references would not work properly and
      the keep.source option failed.  (PR#14459)

    • format.data.frame() now keeps zero character column names.

    • pretty(x) no longer raises an error when x contains solely
      non-finite values. (PR#14468)

    • The plot.TukeyHSD() function now uses a line width of 0.5 for its
      reference lines rather than lwd = 0 (which caused problems for
      some PDF and PostScript viewers).

    • The big.mark argument to prettyNum(), format(), etc. was inserted
      reversed if it was more than one character long.

    • R CMD check failed to check the filenames under man for Windows'
      reserved names.

    • The "Date" and "POSIXt" methods for seq() could overshoot when to
      was supplied and by was specified in months or years.

    • The internal method of untar() now restores hard links as file
      copies rather than symbolic links (which did not work for
      cross-directory links).

    • unzip() did not handle zip files which contained filepaths with
      two or more leading directories which were not in the zipfile and
      did not already exist.  (It is unclear if such zipfiles are valid
      and the third-party C code used did not support them, but
      PR#14462 created one.)

    • combn(n, m) now behaves more regularly for the border case m = 0.
      (PR#14473)

    • The rendering of numbers in plotmath expressions (e.g.
      expression(10^2)) used the current settings for conversion to
      strings rather than setting the defaults, and so could be
      affected by what has been done before.  (PR#14477)

    • The methods of napredict() and naresid() for na.action =
      na.exclude fits did not work correctly in the very rare event
      that every case had been omitted in the fit.  (Reported by Simon
      Wood.)

    • weighted.residuals(drop0=TRUE) returned a vector when the
      residuals were a matrix (e.g. those of class "mlm").  (Reported
      by Bill Dunlap.)

    • Package HTML index files /html/00Index.html were generated
      with a stylesheet reference that was not correct for static
      browsing in libraries.

    • ccf(na.action = na.pass) was not implemented.

    • The parser accepted some incorrect numeric constants, e.g. 20x2.
      (Reported by Olaf Mersmann.)

    • format(*, zero.print) did not always replace the full zero parts.

    • Fixes for subsetting or subassignment of "raster" objects when
      not both i and j are specified.

    • R CMD INSTALL was not always respecting the ZipData: yes field of
      a DESCRIPTION file (although this is frequently incorrectly
      specified for packages with no data or which specify lazy-loading
      of data).

      R CMD INSTALL --use-zip-data was incorrectly implemented as
      --use-zipdata since R 2.9.0.

    • source(file, echo=TRUE) could fail if the file contained #line
      directives.  It now recovers more gracefully, but may still
      display the wrong line if the directive gives incorrect
      information.

    • atan(1i) returned NaN+Infi (rather than 0+Infi) on platforms
      without C99 complex support.

    • library() failed to cache S4 metadata (unlike loadNamespace())
      causing failures in S4-using packages without a namespace (e.g.
      those using reference classes).

    • The function qlogis(lp, log.p=TRUE) no longer prematurely
      overflows to Inf when exp(lp) is close to 1.

    • Updating S4 methods for a group generic function requires
      resetting the methods tables for the members of the group (patch
      contributed by Martin Morgan).

    • In some circumstances (including for package XML), R CMD INSTALL
      installed version-control directories from source packages.

    • Added PROTECT calls to some constructed expressions used in C
      level eval calls.

    • utils:::create.post() (used by bug.report() and help.request())
      failed to quote arguments to the mailer, and so often failed.

    • bug.report() was naive about how to extract maintainer email
      addresses from package descriptions, so would often try mailing
      to incorrect addresses.

    • debugger() could fail to read the environment of a call to a
      function with a ... argument.  (Reported by Charlie Roosen.)    • prettyNum(c(1i, NA), drop0=TRUE) or str(NA_complex_) now work
      correctly.

 

Related Articles

Interview Anne Milley JMP

Here is an interview with Anne Milley, a notable thought leader in the world of analytics. Anne is now Senior Director, Analytical Strategy in Product Marketing for JMP , the leading data visualization software from the SAS Institute.

Ajay-What do you think are the top 5 unique selling points of JMP compared to other statistical software in its category?

Anne-

JMP combines incredible analytic depth and breadth with interactive data visualization, creating a unique environment optimized for discovery and data-driven innovation.

With an extensible framework using JSL (JMP Scripting Language), and integration with SAS, R, and Excel, JMP becomes your analytic hub.

JMP is accessible to all kinds of users. A novice analyst can dig into an interactive report delivered by a custom JMP application. An engineer looking at his own data can use built-in JMP capabilities to discover patterns, and a developer can write code to extend JMP for herself or others.

State-of-the-art DOE capabilities make it easy for anyone to design and analyze efficient experiments to determine which adjustments will yield the greatest gains in quality or process improvement – before costly changes are made.

Not to mention, JMP products are exceptionally well designed and easy to use. See for yourself and check out the free trial at www.jmp.com.

Download a free 30-day trial of JMP.

Ajay- What are the challenges and opportunities of expanding JMP’s market share? Do you see JMP expanding its conferences globally to engage global audiences?

Anne-

We realized solid global growth in 2010. The release of JMP Pro and JMP Clinical last year along with continuing enhancements to the rest of the JMP family of products (JMP and JMP Genomics) should position us well for another good year.

With the growing interest in analytics as a means to sustained value creation, we have the opportunity to help people along their analytic journey – to get started, take the next step, or adopt new paradigms speeding their time to value. The challenge is doing that as fast as we would like.

We are hiring internationally to offer even more events, training and academic programs globally.

Ajay- What are the current and proposed educational and global academic initiatives of JMP? How can we see more JMP in universities across the world (say India- China etc)?

Anne-

We view colleges and universities both as critical incubators of future JMP users and as places where attitudes about data analysis and statistics are formed. We believe that a positive experience in learning statistics makes a person more likely to eventually want and need a product like JMP.

For most students – and particularly for those in applied disciplines of business, engineering and the sciences – the ability to make a statistics course relevant to their primary area of study fosters a positive experience. Fortunately, there is a trend in statistical education toward a more applied, data-driven approach, and JMP provides a very natural environment for both students and researchers.

Its user-friendly navigation, emphasis on data visualization and easy access to the analytics behind the graphics make JMP a compelling alternative to some of our more traditional competitors.

We’ve seen strong growth in the education markets in the last few years, and JMP is now used in nearly half of the top 200 universities in the US.

Internationally, we are at an earlier stage of market development, but we are currently working with both JMP and SAS country offices and their local academic programs to promote JMP. For example, we are working with members of the JMP China office and faculty at several universities in China to support the use of JMP in the development of a master’s curriculum in Applied Statistics there, touched on in this AMSTAT News article.

Ajay- What future trends do you see for 2011 in this market (say top 5)?

Anne-

Growing complexity of data (text, image, audio…) drives the need for more and better visualization and analysis capabilities to make sense of it all.

More “chief analytics officers” are making better use of analytic talent – people are the most important ingredient for success!

JMP has been on the vanguard of 64-bit development, and users are now catching up with us as 64-bit machines become more common.

Users should demand easy-to-use, exploratory and predictive modeling tools as well as robust tools to experiment and learn to help them make the best decisions on an ongoing basis.

All these factors and more fuel the need for the integration of flexible, extensible tools with popular analytic platforms.

Ajay-You enjoy organic gardening as a hobby. How do you think hobbies and unwind time help people be better professionals?

Anne-

I am lucky to work with so many people who view their work as a hobby. They have other interests too, though, some of which are work-related (statistics is relevant everywhere!). Organic gardening helps me put things in perspective and be present in the moment. More than work defines who you are. You can be passionate about your work as well as passionate about other things. I think it’s important to spend some leisure time in ways that bring you joy and contribute to your overall wellbeing and outlook.

Btw, nice interviews over the past several months—I hadn’t kept up, but will check it out more often!

Biography–  Source- http://www.sas.com/knowledge-exchange/business-analytics/biographies.html

  • Anne Milley

    Anne Milley

    Anne Milley is Senior Director of Analytics Strategy at JMP Product Marketing at SAS. Her ties to SAS began with bank failure prediction at Federal Home Loan Bank Dallas and continued at 7-Eleven Inc. She has authored papers and served on committees for F2006, KDD, SIAM, A2010 and several years of SAS’ annual data mining conference. Milley is a contributing faculty member for the International Institute of Analytics. anne.milley@jmp.com

Latex -Lyx -Sweave- Pdf

Lyx (GUI for Latex)

http://wiki.lyx.org/LyX/LyxWithRThroughSweave ( A work of art  in progress!)

http://wiki.lyx.org/LyX/LyX

PDF

http://www.adobe.com/in/products/acrobatpro.html

Latex

MikTex

http://miktex.org/2.9/setup

You use the MiKTeX Net Installer to download all MiKTeX packages and install a complete MiKTeX system. See the section Installing MiKTeX in the MiKTeX manual, for more information.

MiKTeX 2.9 Net Installer
Size: 5.51 MB

Related Articles

PSPP – SPSS 's Open Source Counterpart

A Bold GNU Head
Image via Wikipedia

New Website for Windows Installers for PSPP– try at your own time if you are dedicated to either SPSS or free statistical computing.

http://pspp.awardspace.com/

This page is intended to give a stable root for downloading the PSPP-for-Windows setup from free mirrors.

Highlights of the current PSPP-for-Windows setup
PSPP info:

Current version: Master version = 0.7.6
Release date: See filenames
Information about PSPP: http://www.gnu.org/software/pspp
PSPP Manual: PDF or HTML
(current version will be installed on your PC by the installer package)
Package info:

Windows version: Windows XP and newer
Package Size: 15 Mb
Size on disk: 34 Mb
Technical: MinGW based
Cross-compiled on openSUSE 11.3

Downloads:
There are issues with the latest build. Some users report crashes on their systems on other systems it works fine.

Version Installer for multi-user installation.
Administrator privileges required.
Recommended version.		 Installer for single-user installation.
No administrator privileges required
0.7.6-g38ba1e-blp-build20101116
0.7.5-g805e7e-blp-build20100908
0.7.5-g7803d3-blp-build20100820
0.7.5-g333ac4-blp-build20100727		 PSPP-Master-2010-11-16
PSPP-Master-2010-09-08
PSPP-Master-2010-08-20
PSPP-Master-2010-07-27		 PSPP-Master-single-user-2010-11-16
PSPP-Master-single-user-2010-09-08
PSPP-Master-single-user-2010-08-20
PSPP-Master-single-user-2010-07-27

 

Sources can be found here.

Also see http://en.wikipedia.org/wiki/PSPP

At the user’s choice, statistical output and graphics are done in ASCII, PDF, PostScript or HTML formats. A limited range of statistical graphs can be produced, such as histograms, pie-charts and np-charts.

PSPP can import Gnumeric, OpenDocument and Excel spreadsheets, Postgres databases, comma-separated values– and ASCII-files. It can export files in the SPSS ‘portable’ and ‘system’ file formats and to ASCII files. Some of the libraries used by PSPP can be accessed programmatically; PSPP-Perl provides an interface to the libraries used by PSPP.

and

http://www.gnu.org/software/pspp/

A brief list of some of the features of PSPP follows:

  • Supports over 1 billion cases.
  • Supports over 1 billion variables.
  • Syntax and data files are compatible with SPSS.
  • Choice of terminal or graphical user interface.
  • Choice of text, postscript or html output formats.
  • Inter-operates with Gnumeric, OpenOffice.Org and other free software.
  • Easy data import from spreadsheets, text files and database sources.
  • Fast statistical procedures, even on very large data sets.
  • No license fees.
  • No expiration period.
  • No unethical “end user license agreements”.
  • Fully indexed user manual.
  • Free Software; licensed under GPLv3 or later.
  • Cross platform; Runs on many different computers and many different operating systems.

 

Related Articles