XLSTAT

1
XLSTAT

• Addinsoft

Introduction
2
Outline

• History
• Facts
• Users
• Solution overview
• Use case

Outline
3
XLSTAT History

• 1993 Started development to avoid having to use
  SAS
• 1996 First sales through very basic website
• 2000 Founding of Addinsoft by Thierry Fahmy
• 2001 New interface
• 2002 First modules Pivot 3DPlot integrated
• 2004 Launched a full rewriting project with
  Juergen Jaeger
• 2006 Released version 2006, with VBA interface
  and C engine interface available in 7
  languages
• 2006 First programming interface made available
• 2007 First user conference

History (1)
4
Facts

• MS Excel is the most used spreadsheet software in
  the world (90 market share)
• Excel is the leading numerical data storing
  format
• XLSTAT is the statistical software that offers
  the tightest integration with Excel
• XLSTAT is the most complete statistical software
  for Excel
• XLSTATs market share is still very little
  compared to SPSS or Minitabs, but it is growing
  fast at a 40-50 rate per year
• XLSTAT is modular which allows to lower the cost
  and optimize the usability

Facts
5
Users all around the world

• Users in 96 countries
• Distributors in UK, DE, CA, ES, JP, IL, TW, CN,
  BR,

63 countries
29 countries
Users
6
Users in all industries

• Half users work in companies
• Users in half of the fortune 100 companies
• Most users work in marketing or RD
• More and more students 70 growth (2006-2007ytd)

Users
7
XLSTAT

• XLSTAT is the leading data analysis and
  statistical solution for Microsoft Excel
• The XLSTAT add-in offers a wide variety of
  functions to enhance the analytical capabilities
  of Excel, making it the ideal tool for your
  everyday data analysis and statistics
  requirements.

Why this conference?
8
XLSTAT

• XLSTAT is a Microsoft Excel add-in that has been
  developed since 1993 to enhance the analytical
  capabilities of Excel. XLSTAT relies on Excel for
  the input of data and the display of results, but
  the computations are done using autonomous
  software components. The use of Excel as an
  interface makes XLSTAT a user-friendly and highly
  efficient software. The quality of the
  computations is identical to that offered by the
  leading scientific packages.

8
Paris, June 7, 2007
Why this conference?
9
XLSTAT
9
Paris, June 7, 2007
XLSTAT modules
10
XLSTAT
10
Paris, June 7, 2007
XLSTAT Use case
11
XLSTAT

• An Excel sheet containing both the data and the
  results for use in this tutorial can be
  downloaded by clicking here. The data are from
  Fisher M. (1936). The Use of Multiple
  Measurements in Taxonomic Problems. Annals of
  Eugenics, 7, 179 -188 and correspond to 150 Iris
  flowers, described by four variables (sepal
  length, sepal width, petal length, petal width)
  and their species. Three different species have
  been included in this study setosa, versicolor
  and virginica.

11
Paris, June 7, 2007
XLSTAT Use case
12
XLSTAT

• Our goal is to test if the four descriptive
  variables allow to efficiently predict to which
  species a flower corresponds, and in this case,
  to identify rules that would help classifying the
  flowers on the basis of the four variables.
• Note the same case is treated in the tutorial on
  discriminant analysis.
• After opening XLSTAT, select the XLSTAT/Modeling
  data/Classification and regression trees command,
  or click the corresponding button of the
  "Modeling data" toolbar (see below).

12
Paris, June 7, 2007
XLSTAT Use case
13
XLSTAT

• Once you've clicked the button, the dialog box
  appears. The qualitative dependent variable
  corresponds here to the "Species" variable. The
  quantitative "Explanatory variables" are the four
  descriptive variables. We choose to use the CHAID
  algorithm and we set the maximum tree depth to 3
  to avoid obtaining a too complex tree.

13
Paris, June 7, 2007
XLSTAT Use case
14
XLSTAT

• The computations begin once you have clicked on
  "OK". The results will then be displayed. Below
  the simple statistics for all the selected
  variables, XLSTAT displays information on the
  tree structure. This includes for each node, the
  p-value for the splitting, the number of objects
  at each node, the corresponding the parent and
  son nodes, the split variable, the value(s) or
  intervals of the latter, and the purity that
  indicates what is the of objects that belong to
  the dominating category of the dependent variable
  at this node.

14
Paris, June 7, 2007
XLSTAT Use case
15
XLSTAT
15
Paris, June 7, 2007
XLSTAT Use case
16
XLSTAT
16
Paris, June 7, 2007
XLSTAT Use case
17
XLSTAT

• The following table contains the rules built by
  the algorithm in a less visual but more readable
  way the rules are written in natural language.
  The purity gives the that corresponds to the
  majority category at the node level. The number
  of objects corresponding to the category is also
  displayed.

17
Paris, June 7, 2007
XLSTAT Use case
18
XLSTAT

• Some of the numerous businesses and universities
  currently using XLSTAT
• AXA, BASF, Boeing, BNP-Paribas, BP Amoco, BRGM,
  Bristol-Meyers Squibb Co., Cambridge University,
  Cartier, CIRAD, Champion International,
  ChevronTexaco, Colorado State University, Crédit
  Agricole, Citroën, CNRS, Deutsche Bank Asset
  Management, Dresdner Kleinwort Benson, EDS, Ecole
  des Mines de Paris, Essec, Exxonmobil, Fiat,
  France Telecom, Goldman Sachs, Goodyear, Harvard
  University, IBM, ICI, INRA, K.U.Leuven, Lehman
  Brothers, Leicester university, London School of
  Economics, L'Oréal, Mercer Consulting Group,
  Merck KaG, MIT, Nestlé, Novartis, Ogilvy One,
  Renault, Rhodia, Risk International, Renault,
  Rice University, Roche AG, Sanofi Aventis, Shell,
  Société Générale, Stockholm University, Total,
  UCB Pharma, UCLA, Universidad de Sevilla,
  Università di Firenze, Università di Milano,
  Università di Torino, Université d'Orsay,
  Université de Lausanne, Université de Louvain,
  University of Alberta, University of Bergen,
  University of British Columbia, University of
  Stanford, Young and Rubicam, ...

18
Paris, June 7, 2007
XLSTAT users
19
XLSTAT

• A sample of comments from XLSTAT users
• I have been delighted with the program to date,
  and have found it easy to use, and of immense
  value to my data analysis. It is wonderful not to
  have to change data from one program to another
  to carry out analysis.Paula Cullen, Ireland.
• Let me congratulate you on XLSTAT. It is the
  most impressive statistics add-in for Excel that
  I have ever seen.,Armann I., Canada.
• Thank you for developing XLSTAT. Easy label
  saved me a lot of time and I'm having fun running
  statistical analysis with the other
  tools.,Jeffrey S., UK.
• Fabulous software! I will use it for my
  teaching ,Graham B., USA.

19
Paris, June 7, 2007
XLSTAT citations