STATISTICA Overview
Enterprise Solutions
Web-Based Solutions
Data Mining Solutions
Desktop Solutions
|
New Features and Enhancements
System Level Functionality
Microsoft Office IntegrationStatistics
Graphics
By GroupWorkbooks
ReportsData Filtering, Cleaning, and Handling Outliers
Spreadsheet
Enhancements to WebSTATISTICA Server and Integration of the Desktop and Server VersionsEnhancements to STATISTICA Data MinerDevelopment System/Integration Tool
NEW PRODUCTSSystem Level FunctionalityProjectsYou can save a STATISTICA Project file that will contain the "frozen" state of the current work session, including the locations of all object windows on the screen. Upon opening the project file, the respective analysis can be resumed where it was last saved. Options are provided to save the recorded macro associated with the analysis. Resume Analysis / Rerun Analysis Output that is placed in a workbook will also contain a recorded analysis macro for that output. You can choose to rerun or resume the analysis and can rerun the analysis using a different data set, providing a convenient manner to pick up work from where a project was stopped previously or to perform the same analysis using different data sets. ToolTip improvements
Variable names will break at the nearest character for wrapping if there is no word-break boundary Multiple Version support Multiple, currently licensed versions of STATSTICA (including multiple languages) can be installed on the same computer enabling you to switch back and forth between them as needed. 64-bit support STATISTICA can be installed on 64-bit Windows operating systems. Application Background Window You can select from several standard background styles or can use an image of your choice. Microsoft Office Integration
StatisticsSummaries of analysesSummary, integrated displays with both graphics and statistics available for applicable analyses Descriptives, robust statistics, non-factorial breakdowns, non-factorial tables New statistics including Winsorized means, Trimmed means, Grubb's Test for outliers, Coefficient of Variation, Confidence interval for sample standard deviation, and Percent of valid observations have been added. (See also other tests for outliers in the Data Cleaning and Filtering section below.) T-test, independent by groups Confidence interval for estimates is now available. Correlation matrices Means and standard deviations can now be added to square correlation matrices. Box-Cox Transformation Options to perform Box-Cox transformation have been added, including statistics, histograms, and normal probability plots. The results can be written back to the input spreadsheet. Botched Designs Support for experiments that include "botched" runs (where factors are not set to the specified levels) in 2-level designs (2^(k-p), screening, etc.) Optimal Split Plot Optimal split-plot designs can be generated and analyzed Nonparametrics Limit for nonparametric methods increased to 1-million cases (note that this change has been introduced only to support unusual applications, as more powerful and sensitive parametric tests are preferable to be used with large samples) Multivariate Quality Control The new functionality supported includes: Attribute Agreement Gage Attribute Results now include Fleiss kappa, Kohen kappa, Kendall's coefficient of concordance, and Kendall's correlation coefficient. Graphical results include plots to denote the Assessment agreement by appraiser or by item, with confidence intervals. An assessment disagreement table (that shows how far an appraiser disagrees with the standard) when the response is binary and the standard is known is also available. Capability Analysis - Attribute Gage The results of this analysis includes a summary table and graphs and tests as described in the AIAG Measurement Systems manual Capability ratios for True Position This method evaluates coordinate pairs compared to the "true values" for a center point. Output includes capability ratios and scatterplots for each X/Y pair of variables, with circles and lines that indicate the Tolerance region as specified. Capability Analysis - Gage Capability The variability of gages is compared against the specification limits of the parts. Process Capability ISO/DIN (Time dependent distribution model) Process Capability ISO/DIN (Time dependent distribution models) enable you to compute process capability indices consistent and in compliance with DIN (Deutsche Industrie Norm) 55319 (see Deutsches Institut fuer Normung e.V., 2002) and ISO 21747 (see ISO, 2006). These methods are applicable for situations where consecutive samples of observations are taken from an ongoing production line, and the goal is to estimate the process capability with respect to one or more quality dimensions measured in those samples. Specifically, these standards summarize different distribution "models" for how 1) the observations within each sample can be distributed, 2) how the moments (locations, dispersions, etc.) of consecutive samples can be distributed (e.g., normal, non-normal) over "time," and 3) how best to estimate the process capability based on the resultant distribution (given the distribution of measurements within sample, and across samples/time). Statistics of block data Statistics of block data functions will return the results in a separate output spreadsheet. Top of page GraphicsBag PlotA type of scatter plot using a bivariate generalization of Tukey's univariate box-and-whisker plot to identify distributions (and outliers) in two-dimensional space. Bi Plot A graph used in multivariate process analysis (MSPC) to identify relations between underlying variables and extracted factors to aid in effective root cause analysis. Wafer maps A graphical method for rendering defect data in two-dimensional planes common in the semiconductor industry (wafer plots). Overlaid Contour plots A type of contour plot that identifies "joint-regions" where multiple range requirements for Z are simultaneously satisfied. A graph useful for "visually optimizing" multivariate systems. Overlaid probability plots Probability plots (Normal, Quantile-Quantile, Probability-Probability) can display multiple variables in one graph. Scatterplot with Error Bars A type of scatterplot where points with the same X values are combined and a box/whisker (error bars) are generated at the points Copy graph by dragging Graphs can be copied by dragging. Drag graphs into Word documents Graphs can be placed in STATISTICA documents or external documents such as Word and PowerPoint by dragging. Graph Data as an array Graph data can be accessed in automation as a data array. Locking Graphs Graph files can be locked to prevent users from modifying them. Merging Graphs Graphs can be merged by dragging or copy/paste onto the destination graph. Resolution control for exported graphs You can specifying the resolution (dots per inch) for TIFF, GIF, PNG, JPG, and BMP files when saving a graph via the Save As menu Rotation of objects by CTRL + PAGE UP/PAGE DOWN Graph objects (such as arrows) can be rotated with finer control using Windows-standard CTRL + PAGE UP and CTRL + PAGE DOWN. Scrolling legends A graph legend that is too long to be displayed in the graph window can be scrolled so it can all be seen. The scrolling toolbar buttons enable you to scroll the legend either line-by line, or move all the way to the top/bottom of the legend Selecting items from legend Point markers and fit lines can be selected from the graph legend. Styles and Axis settings Global graphic options control what portion of a graph definition is used when applying a graph style. Top of page By Group
WorkbooksWorkbook multi-item displayThe multi-item display allows for editing of the objects displayed. Save as HTML Workbooks can now be saved as HTML displaying the workbook hierarchy of objects (tree view) in a browser window. Advanced Structure Storage Options allow for more efficient storage of workbooks and provide significantly increased capacity. Locking Workbook User Interface An automation feature that protects a worksheet from inadvertent changes from the user manipulating the workbook tree control. Scrolling, selecting/expanding when dragging When dragging workbook items within the workbook tree, the workbook will automatically expand folders when hovered over and will allow scrolling around the tree by dragging to the bottom or top of the tree. Top of page ReportsNew report options enable you to:
Data Filtering, Cleaning, and Handling OutliersFilter duplicate casesAlso called "de-duping," using selected variables; duplicates (exact matches on all selected variables) can be removed from a data set. Filter Sparse Data Cases or variables that exceed user-specified percentage of missing data can be removed from a data set. Process Relatively Invariant Variables Variables that have a standard deviation below a user-specified threshold level can be removed from a data set. Recode Outliers Using outlier tests such as Categorical, Normal, Grubb's, Percentile, and Tukey, outliers can be recoded to a user-defined value including a specified value, missing data, the mean, a specific percentile, or a boundary value. Process Missing Data Missing data can be recoded to a specific value, mean, or median. Additionally, the variables that have missing data removed can be flagged if the variable contains a threshold percentage of missing data, and if flagged, the entire variable can optionally be removed. k-Nearest Neighbor Interpolation Missing data can be replaced with values that are estimated (imputed) from the data using k-nearest neighbors, substituting the missing data with the "average" value for cases that are similar to - or near - the case in question, with respect to all values that are observed (not missing). Top of page SpreadsheetFast Text Importer User InterfaceFaster options for importing text are now supported. The interface enables you to preview the data, intelligently selects the delimiters and types, and allows each individual variable to be selected in the preview, and the specific type defined. Cell references in Spreadsheet formulas The new formulas can reference individual cells in spreadsheets in the same way as values of individual cells can be referenced in typical spreadsheet applications such as Microsoft Excel. New Spreadsheet formulas New formulas for row-based calculations include:
Options added to treat missing data in code or case ID variables as valid values and to preserve the relative variable order in output. Increase Sort keys to 500 Sort can be initiated using up to 500 sort keys Analysis Case Selection Conditions To prevent common errors, the analysis case selection conditions must now be confirmed before this option becomes active. Text Labels Warning displayed when text values are entered in a spreadsheet and text labels are not enabled for that variable. Missing Data The default missing data code has been changed to -999999998 Merge Improvements Options are now offered to preserve the order of the cases when files are merged, resulting in data in the same relative order as in the first source spreadsheet. JMP 6 files supported Top of page Enhancements to WebSTATISTICA Server and Integration of the Desktop and Server VersionsIntegration with STATISTICA ServerA new user interface is now available in all desktop versions of STATISTICA supporting a seamless integration between with the Server (Web) STATISTICA version. For example, you can now off-load time consuming tasks to the server with a touch of a button, monitor the progress of server computations from the desktop, and easily move STATISTICA objects between the computers. WebSTATISTICA Performance monitors Performance monitor counters can be accessed using standard system tools. Top of page Enhancements to STATISTICA Data Miner
Data Miner Recipe (DMR) A recipe-like step-by-step process to guide you through the data mining process:
Development System/Integration ToolOLE DB Provider for STATISTICA SpreadsheetsStandard SQL can be used to query STATISTICA Spreadsheets from OLE DB compliant applications. Graph Data as an array Graph data can be accessed in automation as a data array. Locking Workbook User Interface An automation feature that protects a worksheet from inadvertent changes from the user manipulating the workbook tree control. Additional supported formulas New formulas for row-based calculations include:
Logging interface An external logging interface that can be accessed from external programs is provided. Additional command line parameters can be used to enable this interface to write out logging information to a report window or an external text file STATISTICA Query completion events Spreadsheets now have new events that get called for attached queries. This is useful for background queries (those you specify with Query.Refresh(True)), though it will be called for background and foreground queries. One function is called when query completes; another is called if there is an error in the query. The events can be hooked up to in a standard fashion from either SVB (using the WithEvents declaration) or from external languages that support connection points. WebSTATISTICA Performance monitors Performance monitor counters can be accessed using standard system tools. Top of page New ProductsSANN - STATISTICA AUTOMATED NEURAL NETWORKSSTATISTICA Automated Neural Networks is a high-performance and easy-to-use application that includes the latest technologies and state-of-the-art algorithms for building and deploying neural network models. Neural network modules include:
Top of page |
||
Features