For the most up-to-date marketing materials on JMP 12 and JMP Pro 12, please visit the new product pages:
New in JMP 12:
http://cqauthor.unx.sas.com:4502/content/jmp/en_us/sandbox/jmp12/jmp/new-in-jmp.html?wcmmode=disabled&previewpath=true
New in JMP Pro 12:
http://cqauthor.unx.sas.com:4502/content/jmp/en_us/sandbox/jmp12/jmp-pro/new-in-jmp-pro.html?wcmmode=disabled&previewpath=true
DRAFT SAS COMPANY CONFIDENTIAL
Current 10-08-2014
New Features in JMP® 12 and JMP® Pro 12 by Application Areas
Data Import, Cleanup and Cleansing
- Query Builder helps users discover interesting data and make correct queries that are repeatable and sharable. Also build custom prompts, sample database and monitor queries in progress through the Query Builder or View > Running Queries.
- Smarter Recode lets you clean up messy data with a suite of powerful tools, including string clustering based on character edit distance. Recode manually or use an automatic recode algorithm.
- Excel Import wizard available on Mac version of JMP.
- Anonymize in the Tables menu is a utility that obfuscates column names and data values while maintaining the data structure.
- File previewer for Mac version of JMP; QuickLook previewer plugin for Mac OSX.
- Close all/multiple windows on the Mac version of JMP. Found in File > Close Multiple.
- New Utilities submenu in Cols menu. Includes Recode, Compress Selected Columns, Text to Columns, Combine Columns, Make Binning Formula.
- New Modeling Utilities submenu in Cols menu. Includes Explore Outliers, Explore Missing Values, (PRO) Make Validation Column, Screen Predictors.
- Support more than 231 rows, 64-bit integers.
Design of Experiments
- (PRO) Covering Arrays, which are used in testing applications where factor interactions may lead to failures. You may want to design an experiment to maximize the probability of finding defect, but minimize cost and time. Covering arrays let you do just that.
- Disallowed combinations wizard available generally in DOE platforms.
- Orthogonal blocking for Definitive Screening Designs.
- Space filling designs support categorical factors.
Statistics, Predictive Modeling and Data Mining
- Visual model editor lets you quickly run and compare various models with a drag-and-drop interface. Supported in many Fit Model personalities.
- (PRO) Generalized Regression, a Fit Model personality, has a new graphical and interactive model selection and informative diagnostics bundle. And it now boasts up to 1,000-times better performance. Additionally, modify tuning parameters and select new distributions such as Cauchy, Beta, Exponential, Zero-Inflated Gamma, Beta Binomial and Zero-Inflated Beta Binomial. Now includes support for effect heredity, inverse prediction and forced effects. Fit models with a stepwise approach by choosing Forward Selection, or perform Quantile Regression by choosing Quantile from the Distribution drop-down list.
- (PRO) Mixed Model, a Fit Model personality, has a new optimizer for greatly improved performance on larger problems.
- Clustering improvements to reduce the time it takes to pre-process data before clustering. The platform now supports stacked data, includes summary features and lets you focus on just a column cluster of interest when there are many columns.
- The Discriminant platform includes many new capabilities, including imputation, robustness, support for missing data, centering canonical variables, increased dimensions, (PRO) support for a validation column and improved scoring functions.
- (PRO) K Nearest Neighbors prediction available in the Partition platform.
- In Oneway and Bivariate, there are now two options for robust fitting of data: Robust Fit, which uses a Huber method, and a Cauchy fit. Cauchy can deal with data sets that include up to 50 percent outliers.
- X11 decomposition available in Time Series platform.
- ANOR (Analysis of ranges) to provide information about group variation added to Analysis of Means methods.
- Informative missing in Fit Model in JMP (from JMP Pro).
- PCA supports wide principal components analysis.
- Cauchy fitting in Response Screening platform.
Data Visualization and Exploratory Data Analysis
- Export JMP reports directly to PowerPoint in a single click, making it easier to build and customize presentations of your analysis, automatically.
- Data tables have image awareness, and support a column of images.
- Many improvements to Graph Builder including: supporting selections in legends; more control of gaps between graph panels; graph matrix; right-clicking gap now supports color, transparency and size; and the graph borders. Also smoothly connected lines, overlaid histograms, new bar element options for needle and float, support for step, horizontal and vertical modes in line element and pages role (a pseudo-BY variable). Finally, a launch dialog is available now for Graph Builder, accessed through Script > Relaunch Analysis or JSL through Graph Builder (Launch Dialog).
- “Make into Data Table” command in Graph Builder. Utilizes expression columns.
- Clickable legends in many platforms.
- Interactive HTML output available for Bubble Plot and Profiler.
- New tiling algorithm in Treemap draws rectangles in order of size while keeping rectangles as close to square as possible.
- Ternary Plot and Mixture Profiler have improved behavior for the magnifier and hand tools, and support editing of Axis properties, including minimum/maximum, major and minor gridlines, and reference lines.
- Selection filtering enables a graph to filter another graph. (Accessible through AppBuilder.)
- New transforms including selected rows transform, distributional transforms (center, standardize, range 0 to 1, Box Cox, Johnson normalizing, informative missing).
- Summary supports “Histogram” statistic.
Quality Engineering, Reliability and Six Sigma
- New Destructive Degradation platform. Choice of numerous model path definitions, transformations and data distribution. Includes a model comparison, CDF and Quantile Profilers.
- Censor code widget consolidation across reliability platforms.
- Grouping variables in Life Distribution.
- Save spec limits in new table – tall: spec limits for each column in rows.
- Able to fit attribute charts to noninteger data in Control Chart Builder.
- New Process capability platform supports “short run” estimates of sigma and “long run” capability. Short run is called Cpk and long run, Ppk.
Consumer/Customer Research
- Multiple Correspondence Analysis, which is like PCA for categorical factors. MCA is often used when a data set has more than two categorical columns or when a data set is multiple answers data.
- Categorical platform improvements, including super categories, a more interactive nonstructured tab and response level toggles.
- Choice models include two more Profilers: multiple choice and probability.
Analytical Application Development
- JSL Mail(); supported on Mac version of JMP.
- As Constant(); surrounds a subexpression by this function to evaluate the expression only once, which can speed up scripts that may have complex but constant expressions.
- AppBuilder allows arranging more than one module in a single window and can be used to edit applications embedded in data tables.
- Display and edit box properties are easier to adjust and customize. Go to Edit > Show Properties to access the properties editor.
- Internet open now gives users access to a web page or a web page as text through JSL. Also creates a script to import images if they are available in an HTML table.
- Get Selected Columns(); now allows an optional “String” argument, just as dt << Get Column Names( String ). This would allow Get Selected Columns(String) to return the following: {"age", "sex", "height"} vs. the current dt << Get Selected Columns(), returning {:age, :sex, :height}.
- Updated "Summarize", "For Each Row", and "Row State" functions to allow the first parameter to be a data table reference. For "Summarize", if the first parameter is not in the form of an assignment, then we assume it is a data table expression. For "For Each Row", if the function call is in the form "For Each Row(x, y)" then we assume the x parameter is a data table expression. For "Row State", if the function call is in the form "Row State(x, y)", then we assume the x parameter is a data table expression.
- JSL support for WrapListBox.
- Script editor – non-contiguous selection.
