JMP Wish List

When asking to display the Confidence intervals in the FDOE platform, only getting a message displayed below, but not the possibility to enter any std error formula.     ... View more

Please consider adding other optimization features in addition to desirability functions. For example Genetic Algorithms. In the quality and process world, we often need to optimize for on target and minimum variations for example.   One intro tool we have used is Crystal Ball and OptQuest addins for Excel. ... View more

Just watched the JMP on Air segment on Statistics Index, The Statistics Index. I like to tell new JMP users about it in JMP classes.   Would like to suggest adding a filter like in the Scripting Index or Preferences. Right now in the Statistics Index you can start typing in the menu to find something, but it is a first character look-up. I search for Steel-Dwass and find it, but if I only remembered something like "Dwass", then it will not show. Or say you want to find any topic on "design", it is scattered throughout the index like "experimental design", "factorial design", "fractional factorial...", etc. ... View more

As a JMP Partner in Brazil, I teach people how to use JMP to solve problems using data, and I’ve received some feedback on the language options available within JMP. While multiple languages are supported, Portuguese is not yet one of them. Including Portuguese would be a significant step toward expanding the JMP community in Latin America. Brazil is a huge market with tremendous potential! Although there are thousands of JMP users here, not all are fluent in English or Spanish. Our competitors’ software includes Portuguese support, which often makes it the preferred choice for some users. Adding a Portuguese option would greatly benefit the JMP community and increase adoption across Brazil. Thanks! ... View more

What inspired this wish list request?  When I create a 2X2X2 factorial design with 2 replicates (16 total runs), and run the reduced model with factor effects A, C, and AC, I expect to get the lack of fit test outcomes because in this case, I have 8 pure error degrees of freedom and remaining 4 degrees of freedom can be used for lack of fit test. But when I really run the model, JMP doesn’t activate the lack of fit report. This is because JMP completely ignores a column B and recognize the model with A, C, and AC as a full model in two factors A and C. But it is truly not a full model in terms of three factors A, B, and C.     What is the improvement you would like to see?    I suggest let JMP keep the information of factors involved in the data (not the model) and use it to provide a LoF report.   Why is this idea important?  For example, suppose a user built an initial model with all possible terms in 3 factors factorial design. By investigating the effect summary window, the user can remove some unnecessary terms one by one. Whenever a term is removed, the user may want to confirm the elimination by checking the LoF report. However, currently, if all the terms involving a specific factor are eliminated, LoF is not provided. This is probably not only restricted to 2-level experimental data; the same thing might happen with any general datasets, I think.       ... View more

What inspired this wish list request?  When creating or importing custom model-based designs, the Compare Designs platform helps comparing and selecting a design based on a model and depending on various metrics and criteria : Power analysis, prediction variance profile and FDS plot, relative estimation efficiencies, aliases matrix and correlations, and optimality efficiencies and average variance of prediction. When creating or importing Space-Filling designs, it may be difficult for practitioners to compare the benefits and tradeoffs between different design types (Fast Flexible, Uniform, Latin Hypercube, ...). MaxPro and discrepancy values are displayed only on each specific design diagnostic panel, so unless scripting a solution (which might be technically complex) to extract these values from several generated designs, there is no workaround to get access to this information and compare several designs. Idea inspired by the topic Get MaxPro value for any Space-Filling design.    What is the improvement you would like to see?  When importing or creating Space-Filling designs, I would like to have a "Compare Designs"-like platform, that enables to link factors from each design to compare them. The comparative evaluation of Space-Filling designs could involve the display of various metrics like MaxPro values, discrepancy and/or Min/Max/Average/Range/StdDev distance between points, which may help compare Space-Filling designs type based on the homogeneity of the points dispersion and the volume/distance covered by points in the design.    Why is this idea important?  Currently, comparing Space-Filling designs is difficult and complex, as there is no built-in platforms for this in JMP. This requires a lot of manual work, from extracting the distance between points (using Hierarchical Clustering and saving the Distance Matrix), to the analysis of the distances. A comparison example is available in JMP Help : Example Comparison of Sphere-Packing, Latin Hypercube, and Uniform Methods (jmp.com), but I would find more useful if the user could generate the comparison easily with any design (created or imported), without the need for scripting or long time (to calculate distance between points and then process the information).    ... View more

What inspired this wish list request?  When creating Custom designs, the Design Explorer platform helps comparing and selecting designs based on various optimality efficiencies and average variance of prediction, depending on different construction choices : runs number, replicates, centre points... When creating Space-Filling designs, it may be difficult for practitioners to understand the benefits and tradeoffs between different design types (Fast Flexible, Uniform, Latin Hypercube, ...) and runs number. MaxPro and discrepancy values are displayed only on each specific design diagnostic panel, so unless scripting a solution (which might be technically complex) to extract these values from several generated designs, there is no workaround to get access to this information and compare several designs. Idea inspired by the topic Get MaxPro value for any Space-Filling design.    What is the improvement you would like to see? When creating Space-Filling designs, I would like to have a "Design Explorer"-like panel option similar to the one in Custom Design, that enables to select several design type and choose various runs number range, and display various metrics like MaxPro values, discrepancy and/or Min/Max/Average/Range/StdDev distance between points, which may help compare Space-Filling designs type based on the homogeneity of the points dispersion and the volume/distance covered by points in the design.    Why is this idea important?  Currently, comparing Space-Filling designs is difficult and complex, as there is no built-in platforms for this in JMP. This requires a lot of manual work, from extracting the distance between points (using Hierarchical Clustering and saving the Distance Matrix), to the analysis of the distances.  A comparison example is available in JMP Help : Example Comparison of Sphere-Packing, Latin Hypercube, and Uniform Methods (jmp.com), but I would find more useful if the user could create the comparison easily, without the need for scripting or long time (to create each design possibility and compare them based on calculated distances).    ... View more

What inspired this wish list request? Confirming that residuals have a normal distribution involves several steps but is fundamental to evaluating the fit.     What is the improvement you would like to see? Overlaid histograms on the LHS of the chart would save a lot of time by providing an immediate view of the normality of the data.     Why is this idea important? It would save some time and remind users to check for normality.     ... View more

What inspired this wish list request?  Unlike other statistical software (Minitab), it seems that JMP doesn't provide detailed formulas for analyses. This lack of details and transparency can make it difficult for users to understand the underlying calculations.     What is the improvement you would like to see?  Provide more formulas for analyses where it makes sense to include them in the support documentation and in cases where formulas are quite complex, simplified formulas for simplified examples so the user can understand the concept.   Why is this idea important?  understand the underlying calculations, it is quite often to get request to explain the analysis, how do you derive this conclusion and what kind of analysis and fundamental concept formulas are being used in the analysis.   ... View more

What inspired this wish list request?  (1) From 'Functional DOE Profiler' feacture in Functional Data Explorer in JMP Pro, I would like to see   (1-1) Confidence Intervals   and   (1-2) Interactive HTML file export from the profiler   (2) From 'Model Comparison' in JMP Pro,    (2-1) 'Model Averaging' (Ensemble) result showed a fixed RSquare (1.00) and RASE (0.00). I think it's a bug.     (2-2) Shelf made 'Ensemble model' that is formulated from other models (Least Squares, Partial Least Squares, SVEM Lasso, XGBoost, Neural Turned with different weight) also showed a fixed RSquare (1.00) and RASE (0.00) value in Model Comparison results.    What is the improvement you would like to see? Why is this idea important?      ... View more

What inspired this wish list request? I came throught the LinkedIn post (and associated publication) from Peter Goos (Model selection for designed experiments with mixed integer programming: MIO and raster plots | LinkedIn), and I found the raster plot a very efficient visualization to help select the most relevant terms in a model.    Here is an example done on the dataset "2x3x4 Factorial" available in JMP Sample Index, where the evaluation metric can be changed (between R², RMSE, AICc and BIC) and is used to order the different models. The raster plot clearly shows where the different models agree or disagree in the inclusion of terms :   What is the improvement you would like to see? If possible in some platforms (for example in Fit Model with Stepwise personality), it would be great to have this kind of visualization to help compare and select the most relevant terms in the model(s). This visualization example can be associated with my previous wish list : https://community.jmp.com/t5/JMP-Wish-List/DoE-Fit-Model-Stepwise-Interactive-visualization-of-models/idi-p/572073 @mia_stephens this wish is similar to the previous one where we discussed about the value of visualizations for models evaluation and selection (link above).   From the Stepwise platform, I used the "All Possible Models" option to generate different models with different terms presence. The rest of the steps can be found in the following script (extracted from the log with some annotations added, sorry for the quality and length) and attached file : // Combine similar TableBoxes into a Data Table (create a datatable based on the output from All Possible Models results) // → Data Table( "Combined Data Table" ) Local( {obj}, obj = Data Table( "2x3x4 Factorial" ) << Fit Model( Y( :Y ), Effects( :X1, :X2, :X3, :X1 * :X2, :X1 * :X3, :X2 * :X3 ), Personality( "Stepwise" ), Run( Rules( "Whole Effects" ), All Possible Models( 8, 50 ) ) ); Report( obj )["All Possible Models", Table Box( 1 )] << Make Combined Data Table; obj << Close Window; ); // Text to columns Data Table( "Combined Data Table" ) << Text to Columns( columns( :Model ), Make Indicator Columns( 1 ), Delimiters( "," ) ); // Group columns Data Table( "Combined Data Table" ) << Group Columns( "X1 etc.", {:X1, :"X1*X2{y-x}"n, :"X1*X2{z-y&x}"n, :"X1*X3{B&A-C}"n, :"X1*X3{B-A}"n, :"X1*X3{D-B&A&C}"n, :"X2{y-x}"n, :"X2{y-x}*X3{B&A-C}"n, :"X2{y-x}*X3{B-A}"n, :"X2{y-x}*X3{D-B&A&C}"n, :"X2{z-y&x}"n, :"X2{z-y&x}*X3{B&A-C}"n, :"X2{z-y&x}*X3{B-A}"n, :"X2{z-y&x}*X3{D-B&A&C}"n, :"X3{B&A-C}"n, :"X3{B-A}"n, :"X3{D-B&A&C}"n} ); // Rename column group: X1 etc. → Terms Data Table( "Combined Data Table" ) << Rename Column Group( "X1 etc.", "Terms" ); // Stack data table // → Data Table( "Visualization table" ) Data Table( "Combined Data Table" ) << Stack( columns( Column Group( "Terms" ) ), Source Label Column( "Terms" ), Stacked Data Column( "Presence" ), Output Table( "Visualization table" ) ); // Change column modeling type: Presence (to facilitate the local data filter in graph builder used after) Data Table( "Visualization table" ):Presence << Set Modeling Type( "Nominal" ); // Recode column: Model ID (to have models displayed by an ID instead of long names) Local( {dt, col1}, dt = Data Table( "Visualization table" ); dt << Begin Data Update; col1 = dt << New Column( dt:Model ); col1 << Set Name( "Model ID" ); dt << Move Selected Columns( {col1}, after( dt:Model ) ); col1 << Set Formula( Recode( :Model, {Num( _rcNow ), Map Value( _rcOrig, {"X1", 1, "X1*X2{y-x}", 2, "X1*X2{y-x},X1*X3{B&A-C}", 3, "X1*X2{y-x},X1*X3{B-A}", 4, "X1*X2{y-x},X1*X3{D-B&A&C}", 5, "X1*X2{y-x},X2{y-x}*X3{B&A-C}", 6, "X1*X2{y-x},X2{y-x}*X3{B-A}", 7, "X1*X2{y-x},X2{y-x}*X3{D-B&A&C}", 8, "X1*X2{y-x},X2{z-y&x}*X3{B&A-C}", 9, "X1*X2{y-x},X2{z-y&x}*X3{B-A}", 10, "X1*X2{y-x},X2{z-y&x}*X3{D-B&A&C}", 11, "X1*X2{z-y&x}", 12, "X1*X2{z-y&x},X1*X2{y-x}", 13, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C}", 14, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X1*X3{B-A}", 15, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{B&A-C}", 16, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{B-A}", 17, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 18, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C}", 19, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{B-A}", 20, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C}", 21, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A}", 22, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{B&A-C}", 23, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{B-A}", 24, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{D-B&A&C}", 25, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{B&A-C}", 26, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{B-A}", 27, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C}", 28, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C}", 29, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C}", 30, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B-A}", 31, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 32, "X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X2{z-y&x}*X3{D-B&A&C}", 33, "X1*X2{z-y&x},X1*X2{y-x},X2{y-x}*X3{B&A-C}", 34, "X1*X2{z-y&x},X1*X2{y-x},X2{y-x}*X3{B-A}", 35, "X1*X2{z-y&x},X1*X2{y-x},X2{y-x}*X3{D-B&A&C}", 36, "X1*X2{z-y&x},X1*X2{y-x},X2{z-y&x}*X3{B&A-C}", 37, "X1*X2{z-y&x},X1*X2{y-x},X2{z-y&x}*X3{B-A}", 38, "X1*X2{z-y&x},X1*X2{y-x},X2{z-y&x}*X3{D-B&A&C}", 39, "X1*X2{z-y&x},X1*X2{y-x},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 40, "X1*X2{z-y&x},X1*X2{y-x},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{D-B&A&C}", 41, "X1*X2{z-y&x},X1*X3{B&A-C}", 42, "X1*X2{z-y&x},X1*X3{B&A-C},X1*X3{B-A}", 43, "X1*X2{z-y&x},X1*X3{B&A-C},X2{y-x}*X3{B&A-C}", 44, "X1*X2{z-y&x},X1*X3{B&A-C},X2{y-x}*X3{B-A}", 45, "X1*X2{z-y&x},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 46, "X1*X2{z-y&x},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C}", 47, "X1*X2{z-y&x},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C}", 48, "X1*X2{z-y&x},X1*X3{B-A}", 49, "X1*X2{z-y&x},X1*X3{B-A},X2{y-x}*X3{B&A-C}", 50, "X1*X2{z-y&x},X1*X3{B-A},X2{y-x}*X3{B-A}", 51, "X1*X2{z-y&x},X1*X3{B-A},X2{y-x}*X3{D-B&A&C}", 52, "X1*X2{z-y&x},X1*X3{B-A},X2{z-y&x}*X3{B&A-C}", 53, "X1*X2{z-y&x},X1*X3{B-A},X2{z-y&x}*X3{B-A}", 54, "X1*X2{z-y&x},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C}", 55, "X1*X2{z-y&x},X1*X3{D-B&A&C}", 56, "X1*X2{z-y&x},X1*X3{D-B&A&C},X1*X3{B&A-C}", 57, "X1*X2{z-y&x},X1*X3{D-B&A&C},X1*X3{B-A}", 58, "X1*X2{z-y&x},X2{y-x}*X3{B&A-C}", 59, "X1*X2{z-y&x},X2{y-x}*X3{B-A}", 60, "X1*X2{z-y&x},X2{y-x}*X3{D-B&A&C}", 61, "X1*X2{z-y&x},X2{z-y&x}*X3{B&A-C}", 62, "X1*X2{z-y&x},X2{z-y&x}*X3{B-A}", 63, "X1*X2{z-y&x},X2{z-y&x}*X3{D-B&A&C}", 64, "X1*X3{B&A-C}", 65, "X1*X3{B&A-C},X1*X3{B-A}", 66, "X1*X3{B&A-C},X2{y-x}*X3{B&A-C}", 67, "X1*X3{B&A-C},X2{y-x}*X3{B-A}", 68, "X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 69, "X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C}", 70, "X1*X3{B&A-C},X2{z-y&x}*X3{B-A}", 71, "X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C}", 72, "X1*X3{B-A}", 73, "X1*X3{D-B&A&C}", 74, "X1*X3{D-B&A&C},X1*X3{B&A-C}", 75, "X1*X3{D-B&A&C},X1*X3{B-A}", 76, "X1,X1*X2{y-x}", 77, "X1,X1*X2{z-y&x}", 78, "X1,X1*X2{z-y&x},X1*X2{y-x}", 79, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C}", 80, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X1*X3{B-A}", 81, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{B&A-C}", 82, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{B-A}", 83, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 84, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C}", 85, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{B-A}", 86, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C}", 87, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A}", 88, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{B&A-C}", 89, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{B-A}", 90, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{y-x}*X3{D-B&A&C}", 91, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{B&A-C}", 92, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{B-A}", 93, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C}", 94, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C}", 95, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C}", 96, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A}", 97, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{B&A-C}", 98, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{B&A-C},X2{y-x}*X3{B-A}", 99, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{B-A}", 100, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{D-B&A&C}", 101, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 102, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{y-x}*X3{D-B&A&C},X2{y-x}*X3{B-A}", 103, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B&A-C}", 104, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B&A-C},X2{y-x}*X3{B&A-C}", 105, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B&A-C},X2{y-x}*X3{B-A}", 106, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 107, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B&A-C},X2{z-y&x}*X3{B-A}", 108, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B-A}", 109, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B-A},X2{y-x}*X3{B&A-C}", 110, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B-A},X2{y-x}*X3{B-A}", 111, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{B-A},X2{y-x}*X3{D-B&A&C}", 112, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C}", 113, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 114, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{B-A}", 115, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{D-B&A&C}", 116, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C},X2{z-y&x}*X3{B&A-C}", 117, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X1*X3{B-A},X2{z-y&x}*X3{D-B&A&C},X2{z-y&x}*X3{B-A}", 118, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{B&A-C}", 119, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{B&A-C},X2{y-x}*X3{B-A}", 120, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{B-A}", 121, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 122, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 123, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{y-x}*X3{D-B&A&C},X2{y-x}*X3{B-A}", 124, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C}", 125, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C},X2{y-x}*X3{B&A-C}", 126, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B&A-C},X2{y-x}*X3{D-B&A&C}", 127, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B-A}", 128, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B-A},X2{y-x}*X3{B&A-C}", 129, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{B-A},X2{y-x}*X3{D-B&A&C}", 130, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C}", 131, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{B&A-C}", 132, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{B-A}", 133, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C},X2{y-x}*X3{D-B&A&C}", 134, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C},X2{z-y&x}*X3{B&A-C}", 135, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B&A-C},X2{z-y&x}*X3{D-B&A&C},X2{z-y&x}*X3{B-A}", 136, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B-A}", 137, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B-A},X2{y-x}*X3{B&A-C}", 138, "X1,X1*X2{z-y&x},X1*X2{y-x},X1*X3{D-B&A&C},X1*X3{B-A},X2{y-x}*X3{B-A}", 139, "X1,X1*X2{z-y&x},X1*X2{y… //Finally, the raster plot! Graph Builder( Size( 936, 645 ), Show Control Panel( 0 ), Variables( X( :Terms ), Y( :Model ID, Order By( :RSquare, Descending, Order Statistic( "Mean" ) ) ), Color( :RSquare ) ), Elements( Heatmap( X, Y, Legend( 10 ) ) ), Local Data Filter( Close Outline( 1 ), Add Filter( columns( :Presence ), Where( :Presence == 1 ) ) ), Column Switcher( :RSquare, {:RSquare, :RMSE, :AICc, :BIC}, Title( "Evaluation metric" ) ), SendToReport( Dispatch( {}, "400", ScaleBox, {Legend Model( 10, Properties( 0, {gradient( {Color Theme( "Stoplight Bad to Good"(1) ), Width( 12 )} )}, Item ID( "RSquare", 1 ) ) )} ), Dispatch( {}, "graph title", TextEditBox, {Set Text( "Raster plot: Model comparison and terms selection" )} ) ) ); And the datatable "Combined Data Table" contains the "Visualization datatable" on which the raster plot is created (also saved as a script if you want to have a look and change the evaluation metric). In order to reduce and optimize the computation of several models (which can be computationally intensive in presence of high number of factors, and possible terms in the models), some safeguards and model creation definitions could be implemented, in the same fashion as the design explorer for DoE, where several designs can be created in parallel. Some options to define could be the highest order of the terms to include in the models, how many different models are created, how much are displayed, ...    Why is this idea important? "All models are wrong, but some are useful" can be a great quote to illustrate the importance of this type of visualization : through the visual comparison of several models ranked by their relevance (R², information criterion, RMSE, ...to be chosen) and their included terms, any user can define which terms to include in the model with more certainty/confidence, or even choose to combine several equally good models. Multiplicity of models is often seen as a concern due to the combinatorial explosion, but this way to summarize and compare visually a lot of models could help see the models' multiplicity as an opportunity for better modeling and decision-making.   @Mark_Bailey and @jacob_rhyne, following our discussion in Discovery Summit on DoE experts expo stand :) ... View more

What inspired this wish list request?  I encountered the situation when trying to realize classification with several Machine Learning algorithms on Mixture Space-Filling DoE data (presentation here : https://community.jmp.com/t5/Online-Abstracts/Synergy-Between-Design-of-Experiments-and-Machine-Learning-for/ec-p/708147#M2). In order to present the results in a visual and clear way, I wanted to graph ternary plots with continuous responses expressing the change in classes, so I opted for the favorite class probability. But I had to do a workaround to actually create the graph, as the Ternary plot requires a formula column, so I did it in two steps : Create a classification model (SVM) and click on "Save Prediction formula". Class prediction is a formula column, but not the calculated probabilities columns for each class. Create a model with the same inputs variables / X's, but the response is the favorite class probabilities column. I can then save the probability formula and use it in the Ternary plot.    What is the improvement you would like to see? As in the model platform the calculation of the class probabilities is already done in order to predict the different classes, I would like to have the calculated class probabilities saved as formula columns in the datatable. This would be a more straightforward option, require less workarounds, and would avoid any model misspecifications in the second step.       Why is this idea important? Ternary plot or Surface Plot are essential tools to better visualize and apprehend RSM/Mixture models behaviors. With categorical responses/classes, it's difficult to represent correctly and continuously the variation in the experimental space, and I think using favorite class calculated probabilities formula could help enhance understand the influence and importance of factors on the categorical response(s).     ... View more

One thing I’ve noticed over time with generating DOE run sheets in JMP is the DOE Dialog script. I like that it’s there so you can easily reproduce the design.   One adjustment I would make is to refrain from actually re-generating the design, which may give a progress bar as well as a new replicate of the run sheet. I think it should take you to the inputs that were being used before the ‘make design’ button is hit because that is ultimately what the user wants to consider (such as trying a new run size or adjusting the model).   It can be distracting to have a new JMP sheet pop up. And in some cases, the progress bar to generate the new design takes a long time, which can be inconvenient as well. This should be a relatively simple upgrade.   ... View more

What inspired this wish list request? When selecting a subset of the candidate set, I needed to select a few runs to force in. I needed a column telling me the run order from the candidate set to guide me.     What is the improvement you would like to see? Add a column of run number from the candidate set as I am choosing runs to force into the new design.     Why is this idea important? Efficiency and avoid errors.     ... View more

What inspired this wish list request?  Need to vizualize in a ternary plot big amounts of mixture data generated from Monte Carlo simulation     What is the improvement you would like to see?  Need in ternary graph same density contour plots as already implemented for other types of graphes    Why is this idea important?  This would greatly improve the vizualization of big data for mixture application     ... View more

What inspired this wish list request?    In chemical industry, it is very common to optimize formulation which results from the selection of raw materials, the ratios in which to blend them and the process conditions used to manufacture them. For example, to optimize a synthesis, you want to choose the reactants, the solvant, the catalyst among a list of possible molecule, to optimize their ratio, and process variable such as temperature, time, mixing power, ...       What is the improvement you would like to see?    Using custom design, it is difficult (but not impossible) to deal with this problematic. For example, you want that the model does not depend on the catalyst molecule when the amount of catalyst is null. To deal with it, we need to use some tricks such as the one describe here : https://community.jmp.com/t5/Discussions/How-do-I-indicate-that-a-mixture-factor-can-take-a-number-of/m-p/387460#M63775     Several approach exists in the litterature : - Muteki, K., MacGregor, J.F., Ueda, T., 2007. Mixture designs and models for the simultaneous selection of ingredients and their ratios. Chemometrics and Intelligent Laboratory Systems. - Hada, S., n.d. Product and Mixture Design in Latent Variable Space by Chemometric Techniques. It would be very interesting to have a plateform in JMP that can implement this kind of DoE.     Why is this idea important?    This is very important to speed up the formulation of many products such as tire, shampoo, polymers, flame retardant, and so on... In our company, this kind of DoE represent a least one third of the DoE. ... View more

I don't have a lot of faith in the 'plot residuals by normal quantiles' for doing model diagnostics. I have typically saved the residuals from the model to a new column, then used the Distribution platform to create a normal quantile plot there.   The fit of the solid red line does not seem to match up well with where the residuals fall using Fit Least Squares. The slope of the solid red line is not a good fit. I looked at a few sample data sets and saw issues with the default model scripts in these data sets:   2x3x4 factorial Bounce data Wine data Below is a plot of the 2x3x4 factorial Residual Normal Quantile Plot. I've been told that the slope of the solid red line is based on the RMSE of 2.02.     If I save the residuals to a new column and regenerate the Normal Quantile Plot from the Distribution plot, the line is based on the standard deviation of the Residual Y column of 1.03. It looks like a much more sensible fit from a model diagnostics perspective.   Is this something that can be updated in future versions of JMP? Note that when you use Fit Mixed, the results in the Residual Plots there match my desired result from manually saving the residuals to a new column. ... View more

What inspired this wish list request?  I wanted to compute the uncertainty of prediction fastly and easily for predictive model without using bagging. Indeed, bagging is quite slow to compute with the profiler.   What is the improvement you would like to see?    I would like to save the all the trees formula when I use the Bootstrap Forest Plateform. Or at least the standard deviation of the distribution.     Why is this idea important?    This is very important for me to be able to compute acquisition function when I do Surrogate Model Optimization (Bayesian Optimization). I tried to use bagging with the partition plateform, but the result is not satisfactory. The gaussien process plateform do not deal with categorical variable.     ... View more

What inspired this wish list request? I have been working my way through the excellent Visual Six Sigma: Making Data Analysis Lean (Wiley and SAS Business Series) (2016) book by Cox Gaudard and Stephens from SAS Institute.   I decided to test out the new Easy DOE platform on the example for reproducing the Anodize_CustomDesign_Results.jmp data.  (I had recently been able to reproduce the results with JMP step by step.)  This was quite easy and proved the power of the Easy DOE platform, both very visual and versatile with Guided Mode and Flexible Mode (it feels a bit like TurboTax modes).   Everything went well till the 6th of 7 steps in Easy DOE which is the Predict Step.  In this step, the profiler should show the grid for the effect of the 5 factors on the 4 outcome variables, plus the desirability row and column.    The 4 desirability row and 5 desirability column grid results are missing from the output even though the option is turned on in the red triangle.  When I tried to run the option of Maximize Desirability, this did indeed work well and gave the expected result.   But as I said previously the graphical output of the Desirability is missing from the grid.  (I am running the latest version of JMP 17.2.0)     What is the improvement you would like to see?  That not only can I maximize desirability, but also see the Desirability Functions as I already can outside of Easy DOE.   If you consider this suggestion makes the Guided Mode too busy visually, at least show the Desirability Functions in the grid in the (more advanced) Flexible Mode.     Why is this idea important? One of the most useful features of the Profiler is visualizing the Desirability Functions.  They permit visual sensitivity analysis, suggest trade-offs among (nearly) optimal solutions, etc. So I think that practitioners and students learning JMP can greatly benefit.     ... View more