Statistica



TULSA, OK, USA [May 7, 2013] – StatSoft, Inc., has released STATISTICA Version 12, the next major step in the evolution of its high-performance predictive modeling and data analytics reporting platform.

Building upon the foundations that have already earned STATISTICA more awards than any other data analysis product in the industry, StatSoft has responded to users’ feedback on the evolving business analytics landscape by enhancing or adding powerful new tools that will bring Big Data processing out of the tech room and into the boardroom.

After starting the Statistica application, type F1. Look at the bottom of Contents tab for Statistica Object Model. Virtually every detail of data importing, visualization, cleaning, data analysis, predictive analytics, and reporting can be controlled via a scripting language. Macros can be recorded within Statistica and saved for re-running later. Statistica has a wide selection of scientific and technical charts featuring built-in analytic options. There are 2- and 3-dimensional graphical displays, ternary graphs, special 4-dimensional graphs, multidimensional graphs, categorized multigraphs, matrices of graphs, icons, tessellations, spectral 2- and 3-dimensional graphs, compound graphs.

“The release of Version 12 is another milestone for us,” noted Dr. Thomas Hill, StatSoft’s VP of Analytic Solutions. “Extensive feedback from our clients about their emerging needs in this hyper-competitive environment has allowed us to get ahead of the curve with V12, to deliver superior efficiency and capability to our clients across automated manufacturing, financial services, insurance, and other domains. This release will further bolster our status as the industry leader for enterprise analytic BI solutions for Big Data.”

Version 12 has been designed to make the user experience more enjoyable and productive, with new features that make model-building and data comprehension more intuitive and accessible to business people with little or no advanced analytics experience.

STATISTICA 12 represents a major technological step for the STATISTICA Enterprise™ platform,” stated George Butler, StatSoft’s VP of Platform Development. “New STATISTICA Enterprise features and enhancements make creating powerful analytic workflows easy, even for novice users. The STATISTICA Decisioning Platform® provides a foundation for any industry interested in data modeling and shortening its model development/deployment lifecycle.”

STATISTICA 12 includes a re-invented visual workflow, Big Data performance enhancements, new analytics, and expanded capabilities for enterprise-level, multi-document storage.

Additionally, STATISTICA 12 introduces a variety of practical nodes, query builders, and reporting table tools designed to move STATISTICA solutions into the hands of a broader base of business users while still appealing to power users already familiar with StatSoft’s flagship product.

Top new features include:

  • Big Data performance boost across the entire product line for faster processing of all data volumes, especially relevant to the predictive and prescriptive capabilities of Decisioning Platform®
  • New HP (High Performance) Analytics version of STATISTICA Enterprise supports in-memory and parallel processing of massive data sets
  • Enhanced visual user interface in the new Advanced Query Builder means that novices and experienced users alike can write even complex SQL queries
  • Reinvented visual analytic workspace in STATISTICA Enterprise™ and Data Miner products provides a more intuitive user experience, with greater visual workflow and storage capabilities to help users understand and communicate their findings
  • A new product, the highly flexible STATISICA Reporting Tables, enables users to visually build tables of summary statistics and use these tables in presentations and other reports
  • New Data Health Check node helps identify potential data input problems and facilitates data cleaning for submission to analyses; the tool is optimized for Big Data sets
  • New spreadsheet file formats, virtual variables, and importing options result in measurably faster software performance

ABOUT STATSOFT, INC.

StatSoft was founded in 1984 and is now one of the world’s largest providers of analytics software, with 30 offices around the globe and more than one million users of its STATISTICA software. StatSoft’s solutions enjoy an extremely high level of user satisfaction across industries, as demonstrated in the unprecedented record of top ratings in practically all published reviews and large, independent surveys of analytics users worldwide. With its comprehensive suite of STATISTICA solutions for a wide variety of industries, StatSoft is a trusted partner of the world’s largest organizations and businesses (including most of the Fortune 500 companies), providing mission-critical applications that help them increase productivity, control risk, reduce waste, streamline operations, achieve regulatory compliance, and protect the environment.


A brief tutorial

The tutorial is a brief look at what Statistica for Windows iscapable of doing. Examples will come from StatisticalMethods for Psychology by David C. Howell. It isnot our intention to teach you about statistics in this tutorial.For such information, you should rely on your classes instatistics and/or a good textbook. If you're a novice thistutorial should give you a feel for the programme and how tonavigate through the many options. Beyond that, the StatisticaHelp Files represent all of the documentation relevant to usingthe programme.

Statistica

The Basics

StatisticaInteracting with Statistica is essentially like working with anyother programme designed for Windows. However, there are somedifferences that will be discussed as we progress. Virtuallyany statistic that you wish to perform can be accomplished in combination with pointing and clicking on the menus and variousinteractive dialog boxes. You may have noted that the examplesin the Howell textbook are performed/analyzed via code/programminglanguages. That is, Statistica, like many other packages, canbe accessed by programming short scripts, instead of pointingand clicking. We will not cover any programming in this tutorial.However, this should not stop you from exploring this avenue viathe online documentation and examples.

Presumeably, Statistica is already installed on your computer.If you don't have a shortcut on your desktop go to the[Start=>Programs] menu and start the package by clickingon the Statistica icon. If you don't know what I'm talking about,ask some one at the local Computer HelpDesk.

Statistica

Before proceeding any further I should say a few words abouta very simple convention that will be used throughout this tutorial.In this point and click environment one must often navigate throughmany layers of menu items before encountering the required option.In the above paragraph the prescribed task was to locate the Statisticaicon in the [Start] structure. To get at that icon, one mustfirst click on [Start], then move the pointer to the[Programs] options, before locating the Statistica icon.This sequence of events can be conveyed by typing [Start=>Programs]. That is, one must move from the outer layer of themenu structure to some inner layer in sequence...

Now, back to the tutorial.

Once you have double clicked on the Statistica icon a new windowwill appear on the screen. This window will provide you withmenu with several choices for analyses, like the following,


Using the scrollbar, you will note that your choices cover thefull range of statistical procedures that you are likelyto encounter. By double clicking on one of the options a newwindow will appear with a spreadsheet like interface. For now,double click on the Basic Statistics/Tables option to produce the following window,

Statistical Process Control


[Note: I'm not fond of having a fullscreen display. The initialdisplay of the above window occupied the fullscreen. To reducethe size of the window, click on the middle button on the upperright corner of the display. That is, the button with the overlappingboxes. I imagine there is preference setting that would automaticallylaunch Statistica as a windowed display....]

The above window represents the interface to the Descriptive Statistics module. I suspect that this modular approach maybe rooted in using less resources for a given set of analyses.That is, by just loading the resources for the required statistics,the load on the system is reduced. Or, I could be mistaken --doesn't matter. Once the programme has been started one canopen another module to conduct other analyses. For example,by clicking on [Analysis=>Other Statistics] you will once again bepresented with the original choice menu. This time around, doubleclick on the [ANOVA/MANOVA] menu item, or click on the End & Switch] button. The first approach will result in anew window, in addition to the Basic Statistics window, on thescreen. So, one can have many instances of Statistica openedat the same time. The latter approach will close the Basic Statisticwindow, then open the ANOVA/MANOVA window. This should reduce the load on memory resources, howevever, the final choice is up toyou.

As you can see, there are a number of menu options relating to statistics.There are also shortcut icons on the vertical and horizontal toolbar.These serve as quick access to often used options. Holding yourmouse over one of these icons for a second or two will produce a shortfunction description for that icon. The current display is thatof an empty data sheet. Cleary, data can either be entered manually,or it can be read from an existing data file.

Browsing the File menu, displayed below, reveals nothing surprising - many of the options are familiar. Although, thedetails are specific to Statistica. For example, the [New]option leads to the opening of a new data sheet, with a uniquename. Other interesting options include [Open Data] and[Import Data].

Statistica


Statistical Methods

The [Open Data] option will launch a dialog box that canbe used to open existing data files in a native Statistica format.Like other application packages (e.g., WordPerfect, Excel, ...)Statistica also has it's own format for saving data. In this case,the accepted extension for any file saved using the proprietaryformat is 'sta'. So, one can have a datafile saved as 'data1.sta'.Anyways, the format is not readable with a text editor (e.g.,Notepad), it is a binary format. The benefits are that all formattingchanges are maintained and the file can be read faster, hence the[Open Data] option. It is specifically meant for files savedin the SPSS format. The second option, [Import Data], asthe name suggests is to read files that are not in the statisticaformat. These include ASCII, or text, datafiles and variousspreadsheet formats. By clicking on [Import Data => Quick]one can specify the format of the data file to be read. That is,a new dialog box will appear, and one can scroll through thelist of available file types - explore this dialog box.


Reading datafiles that are in a spreadsheet format requiresvery little effort. The process is seamless, however, readingASCII files requires that the user know something about theformat of the datafile. Otherwise, one is likely to getstuck in the process of reading. There are a number of acceptableformats - comma separated, space separated, semicolon separated,tab separated, and even a user defined format. For example, thefollowing is comma separated file with 3 variables and 6 cases.

You should experiment with some artificial datafiles. It's possibleto use a simple editor like NotePad - found on all Win95 machines -or a wordprocessor. If you use a wordprocessor, make sure you savethe file in text format.Once you have a simple ASCII datafile click on [Import Data => Quick], and locate the file. After you have double clicked onthe example file you'll be asked to specify the format, andthat should do the trick.

Before we move onto actual data, click on [Analysis].The menu that appears reveals a number of basic statisticsthat can be calculated. Recall, that we're currently usingthe [Basic Statistics] module. The components of the[Analysis] menu will change depending on the modulebeing used.


Statistical Scale