JMP Wish List

Can you ,please, add a new feature of 3D contour plot type in which multiple layers of 2D colour coded contour plots combined into a 3D cube that can be rotated with a slider, transparency control and data filter to represent and visualize more complex relations among 4 variables directly in one plot. ... View more

Not sure if this is the best place to bring it up, but something that has bothered me for years is the "marker size" option in Graphbuilder. Even if I use the maximum size (level 6), it's typically too small for a publication, though I just realized you can choose "other" and go as large as you want. Although this gets me to where I want to be, I wonder if it would be possible to, rather than present size levels 1-6 as options, instead have a bar that you can drag to set the icon size precisely (in which you can watch the change in real time, i.e., dynamic resizing). Another alternative could be to treat the icons as fonts, in which case you could use the font editor to increase their size to a specific font size. This might be good for people like me who paste or export these into Adobe Illustrator or other applications, in which case you may need a specific font type and size.    ... View more

Hello,   Is it possible to add multiple responses profiler on an interactive HTML?  It only works with a single response in my example.   In the attached you can run the embedded model (Desired Profile) and the profile works correctly.  If I save as an Interactive HTML, it does not work.  I can manipulate the table to get a "working" result but it's not the desired result.   Thanks, Sean ... View more

When using violin plots (density), there is an option to have equal widths per variable.   However, we are unable to plot the dot distribution with the same shape as it does not respect the width of each variable.     // Start of script; Names Default To Here( 1 ); Clear Log(); dt = Open( "$SAMPLE_DATA/Bands Data.JMP" ); Data Table( "Bands Data" ) << Stack( columns( :proof cut, :viscosity, :blade pressure ), Source Label Column( "Label" ), Stacked Data Column( "Data" ), Drop All Other Columns( 1 ) ); Graph Builder( Size( 534, 456 ), Show Control Panel( 0 ), Variables( X( :Data ), Y( :Label ) ), Elements( Points( X, Y, Legend( 5 ), Jitter( "Packed" ), Jitter Limit( 0 ) ), Contour( X, Y, Legend( 6 ), Violin Scaling( "Equal Width" ) ) ) ) It should be straight forward to achieve something like this but for a number of variables.   ... View more

It is really nice that you can colorize the prediction profiler. It sometime happen that I used that option and wanted to have the profiler without the colors for an PPTX later. Since I did not found any hint to decolorize every part of the profiler at once it is very hard to decolorize a profiler with 4 parameters and 15 responses by clicking with the mouse and select white as background space. Could you please include that function in the next Version of JMP?   Best regards, Christian ... View more

For a structured regression model with mediation in the SEM platform, can JMP compute the indirect effects and its standard errors (using normal theory or the delta method) and display as a table under Parameter Estimates?  It will be super to show this also in the prediction profiler (if possible).   Thanks ... View more

I really like the graphlets in the MDMVCC platform, especially when going from the T2 on the principle components to the Diagnosis of the process plots.   Based on some of the other subroutines in the multivariate platform, and combined with the score plot, I think it would be very helpful to have a parallel plot as a subroutine in the MDMVCC. This could very much help to see the connections between certain processes and how they might interact to lead to OOS or OOL processes.   Just a thought.   Thanks!, DS     ... View more

I want to have percentile-based whiskers in Graph Builder - box plot. This also applies to box plots on Distribution platform.   JMP's box plot does not allow the user to specify the location (length) of whiskers and only offer "Tukey" method. (Oddly this is not described in JMP's help page on Graph Builder - Box Plot, but it is in JMP Clinical's page below.) https://www.jmp.com/support/downloads/JMPC71_documentation/Content/JMPCUserGuide/GR_C_0003.htm   It doesn't serve my purpose because my data contains dirty outliers and Tukey whiskers, with or without dots for outliers, draw attention to these extreme values. Instead, I draw box plots with whiskers at 10 and 90 percentiles, 5 and 95, 2.5 and 97.5, 1 and 99, etc. For this purpose I use some other app, like Graphpad Prism. Below is a link to Graphpad Prism's help page on this. https://www.graphpad.com/guides/prism/8/user-guide/box-and-whiskers.htm I would like JMP to have such capability.   I believe we need to update the way we use box plots as it doesn't seem to go extinct in near future even with new and richer graph types available, partly because its elements correspond to important summary statistics. ... View more

Match JMP probit analysis with PROC PROBIT from SAS, including LACKFIT option. SAS will assume normal or T distribution depending on the lack of fit test result. LoD and CI results from the two platforms can vary based on this difference. ... View more

Has anyone working on disposable medical products (or other single use products) ever wished that JMP had a user-friendly graphically driven "Reliability Plotting" platform or similar graphically driven method for calculating Reliability and Confidence (R&C) directly for non-normal data?    Reliability Plotting is a method originally developed with industrial application in electronics, and is a way of applying transformations to univariate data on a plot of Median Rank (or % Cumulative of the Data, transformed if necessary) versus the Raw data (or a transformation of that data, if necessary) -- in order to linearize that plot for the purposes of R&C estimation (where R in this context refers to % of product in-specification and C refers to % confidence level).           So, ultimately transformations are applied in Y and/or in X to get a best-fitting straight line, and the % in-spec can be determined at the extrapolated value of Y at the corresponding specification limit(s) in X.  From this extrapolation a confidence limit can be calculated, and the resultant %in-spec (i.e. % Reliability) at the chosen % Confidence level can be obtained.   This method is very powerful because it uses more information than k-tables for computing C&R, where it considers the relative location of the data to the line of best fit, whereas the classical "k-tables method" uses only the mean and standard deviation of the data.    I attach an infographic to further explain conceptually what I am talking about! (courtesy of my friend John Zorich, MS, CQE, who developed his own Excel spreadsheet version of this method, with many textbook references).     JMP has the "Life Distribution" Platform, and "Custom Estimation" and these tools provide some semblance of the Reliability Plotting method, but bottom line is they are not designed to output the C/R in a direct, intuitive, and graphically-driven way.  I would love to know if anyone else is familiar with this technique and if there others would find it useful to have a ready implementation in JMP?   Thanks! I include @Daniel_Valente at JMP whom I've discussed this with in recent weeks.    Cheers, @PatrickGiuliano    ... View more

Given a FROM and TO column of data, JMP should be able to make a state diagram.   So for Data like this: FROM            TO A                     B B                     C C                     A C                     D D                     A     The State Model would be:       _____________________     /                           \                 \  |                              |                  | V                              |                  | A   -->    B     -->      C     -->       D   If JMP could do this with graphical output that would be amazing! Then being able to add graphlets when I hover of them I can monitor my whole process graphically! Thanks!     ... View more

  Graph Builder Continuous Color Legend Horizontal    Please allow for turning the color bar of a continuous variable in Graph Builder horizontal, especially when the legend is placed on the top or bottom.   Thank you ... View more

Problem statement: I will often plot a time trend, setting a categorical variable to Overlay role (and often checking several different factors, so first overlaying by column A, then overlay by column B, then column C, etc.) But when there are categories containing many different groups, the plots can get extremely busy (see below,) and it's hard to tell "large N" smoother lines that have a large data set in that group from "small N" lines that are just a spline fit to ~3-5 data points.   Suggestion: Could a "dynamic width" option be added to Smoother that adjusts the line thickness depending on data volume? Lines with a large number of data points would be thicker, and lines with only a few data points would be very thin.   Ideally, the data "density" could also be taken into account: if there are only 2 data points from group X in 2015, but 30 data points in 2016 and 100 in 2017, the group X smoother line thickness would be thin in 2015 and thicker in 2017. (Basically the line thickness would be inversely proportional to the time delta between points - but smoothed so the thickness of the line doesn't become infinite if the time delta between two points gets very close to 0.)   Example problem: This is the kind of "busy plot" I'd like to use this feature with: if the smoother lines with lots of data were very thick/bold and the smoother lines with just a handful of data points were very thin, it would be much easier to see what is a real signal. (There are real signals in this data, I promise.) I hope this is clear enough, please let me know if any clarification is needed. ... View more

In some case, it is needed to manage more than one data set. Could be useful to have the possibility to have in the same graph the different data distributions and for each of them have the posssibility to generate the dedicated fitted line (by the single selection in the continous/discrete fit).   Thanks in advance for your attention.   Best Regards, Simone Cianchini ... View more

I would like the ability to record and save an animated gif when using the column switcher on mac similar to what you can do on Windows. ... View more

Can you please fix the HTML interactivity in DOE model's profilers when one or a few constraints are defined? My DOE model’s profilers in the HTML file became static when I added the constraints. They were interactive when I had no constraints. Unfortunately, including constraints in most DOE models is inevitable.   ... View more

When we perform a regression (simple, multiple, DOE,...) one key activity (for the model check) is the evaluation of the residuals. Typically, the residual evaluation consist in 4 key checks: 1) Normality of the residuals distribution (with the distribution and p-value evaluation); 2) Randomly distributed (typically we use a chart with residuals versus the fitted values); 3) No residuals trend (typically we use a chart with residuals versus the order of the data); 4) Standardized residuals between -2 and +2.   JMP can manage this kind of residuals... but in different menu and in some case not so user-friendly. One improvement opportunity: create a button "Residuals Check" -> when the user click on it -> JMP generate one single screen (one page dashboard) with 4 quadrants that reppresent the 4 different residuals check. In this way... this is more easy for not expert users and very fast verification in one click. Thanks in advance for your attention and feedback. Best Regards, Simone ... View more

This method an effective way to aggregate multiple data types on a genome wide scale. Similarity Network Fusion uses many views of a network and fuses them to build an overall improved matrix. The learned status matrix can improve clustering, and classification.   Reference: Similarity network fusion for aggregating data types on a genomic scale. Nature Methods; 2014 Mar;11(3):333-7. https://cran.r-project.org/web/packages/SNFtool/index.html https://github.com/maxconway/SNFtool ... View more

When we have a distribution (normal or not normal), we can ask to JMP to calulate the CPK. The output is "one specific number". Could be useful to have not a number but a range (see the attached example). Thanks for your attention. Best Regards, Simone     ... View more