JMP Wish List

What inspired this wish list request?  As a reliability engineer, I frequently use the Life Distribution platform for its distribution plotting and fitting capabilities.  These plots are important not only for internal topics but also for critical customer communications.  Unfortunately, the method of plotting data points in this platform changed in JMP 17 in a way that is less desirable.   These plots are often used to demonstrate the quality of fit of a distribution model to data.  Such demonstrations can be crucial in obtaining acceptance.  The Kaplan-Meier / Turnbull estimates used in JMP 16 were appropriate for this, as a good MLE distribution fit often came close to the estimates.  For example, see the Weibull MLE fits below (same data table, different versions).  Both MLE Weibull fits are identical, however JMP 17 points are significantly below the fit.  This change to the default behavior is troubling - now I must question which version of JMP was used when reviewing plots.   JMP 16 – “great fit!” JMP 17 – “not such a great fit?”  (same fit!) In JMP 17 the user can select Points (black, default) / Step Function (red) / Both / None.  Both selected below: However there are a some issues with this: User has no customization options for the red markers (which are what should be default).  Need ability to adjust color, marker, size, etc. just like regular markers. Buggy behavior: selecting Step Function does not work when going from one of the other options.  It works when the platform is first launched if selected as platform default in preferences, however when toggling from Points back to Step Function it once again shows nothing.  Even if selected as default in preferences, once the “Nonparametric” distribution is deselected the red markers disappear and can only be recalled if “Both” selected.     What is the improvement you would like to see?  The default point positions should be the “step function” points (Kaplan-Meier / Turnbull estimates) to be consistent with previous versions.  It would still be nice to be able to select midpoints, however each option should have full marker customizability.      Why is this idea important?  These plots are often used to demonstrate the quality of fit of a maximum likelihood distribution model to data.  Such demonstrations can be crucial in obtaining acceptance.  The previous default Kaplan-Meier / Turnbull estimates were appropriate for this, as a good MLE distribution fit often came close to the estimates.  Deviating from this now with JMP 17 causes confusion and is inconsistent with previous versions.       ... View more

What inspired this wish list request? I have a table that is already sorted in a specific manner and I can't resort it. I need to do a Column Statistic, but not a standard statistic. I can build JSL in the column formula to go through the whole table to create a matrix "by" the categories of interest, but for this larger table, the JSL layer would be pretty slow and inefficient. JMPs canned columns statistics already gather these matricies - and quite fast - so I would like to leverage that.     What is the improvement you would like to see? I would like to see a column function that is more generic... so like ColMean(column, by1, by2, by3) returns the mean of a matrix of values by 'by1', 'by2' and 'by3' .... I would like a function that just returns the matrix of values prior to any math being done.     Why is this idea important?  As mentioned in the first section, coding this in the column formula would be very inefficient. Coding this in a separate script (a table script, perhaps) could be faster, but then not as 'live' as a formula in a column. With JMPs base infrastructure already existing, being able to leverage it further would be ideal.     ... View more

What inspired this wish list request?    Multivariate platform's Outlier Analysis allows you to save Mahalanobis Distances, Jackknife Distances, etc to a column.  It will plot these and show a calculated Upper control limit on those for current alpha - very useful for using the distances to assess whether a point is an outlier.  JMP puts a single value (ie Jackknife Value) into the column properties (ie on the Jackknife Distances column) that is this UCL.  However, if there was a By group, there are distinct UCL values for every group (based on number of non-missing points in the group) and this single value in the column properties appears to just be for the last group.     One might think you could extract them from the platform's report but it doesn't seem like it - they are in Frameboxes, not textboxes in the report and I don't see a way to extract those values.       What is the improvement you would like to see?    In my use case, with a By Group I'd like the UCLs in a column as well.  But what I'd like to see is maybe an ability to ask the platform for those, maybe similar to the way you can ask a fit in Fit Model for various calculated values using  <<GetEstimates, and get a list of those for all the by groups.   Related: it would be nice if in cases like this there were a good way to <<Get what the groups are in order to use that list of values better? At least I'm not sure how to do that without parsing it out of the outlinebox names on the report.  Would be nice to get a handy list in same order as the values in what you get from <<getestimates so you know for sure what group each is for.  It has come to my attention that different platforms vary in terms of what groups they will include in their report (or skip due to inadequate data) and the order groups will be in may not be as obvious as it used to be now that JMP will user "numerical ordering" in some cases.     Why is this idea important?    I think it's clear why the UCLs are useful - it's what you'd compare the distance to in order to see if it's a large distance, for a given confidence?   Here's a community post where someone wanted this value, they just weren't using a By I guess because were happy with the single value in column properties:Solved: Jackknife Predicted Values - JMP User Community       ... View more

☐ cool new feature ☐ could help many users! ☑ feels like a „bug“ ☐ nice to have ☐ nobody needs it   What inspired this wish list request?  Some display boxes are not supported via Application Builder: https://community.jmp.com/t5/Discussions/range-slider-box/m-p/696994/highlight/true#M88219    What is the improvement you would like to see?  Please add Application Builder support for the missing Display Boxes:   Range Slider Box (according to Jmp Support, TS-00074741, this was already requested before and is already on the radar of Jmp development) other missing Boxes, which may not be on Jmp dev. radar: Web Browser Box Scene Display List Calendar Box ...   Why is this idea important?  Application Builder is very helpful for users who design their own Reports. It should support all the functionality of Jmp.         more wishes submitted by   ... View more

What inspired this wish list request?  JMP17 Fit Y by X, "Means and Std Dev" output report changed the standard columns output. Columns can be changed manually by right-clicking on the table and choosing the Columns I want to see. Preferences to make these Column selections does not exist.  We do not want to create a script or distribute an Add-In.  We would prefer to easily update the Preferences.   What is the improvement you would like to see? Column selection from this output table should be available in the Platform Preferences pick-list.     Why is this idea important? We have several engineers in our company who use the "Means and Std Dev" report in their standard reporting for their important parameters.  They do not want the additional columns in the new standard report.  They want the ability to customize their own preferences for this platform.     ... View more

☐ cool new feature ☑ could help many users! ☑ removes a „bug“ ☐ nice to have ☐ nobody needs it   #myTop10_2022     Get rows where is extremely slow if the column is referenced with a varible (here i=5 ) dt = Open( "$SAMPLE_DATA/Wafer Stacked.jmp" ); t0=hptime(); i = 5; dt << get rows where( As Column( dt, i ) == 10 ); (hptime()-t0)/1000000; instead of a direct reference like in dt << get rows where( As Column( dt, 5 ) == 10 ); My wishes: Is it possible to "fix this issue in the background"? e.g. by a setting in the preferences: "hereby I confirm that JMP is allows to pre-evaluate variables to speed up code execution - accepting that this could lead to unintended/unusual behavior" (I am not good enough in JSL programming to predict all the collateral effects - would such a mode make sense?) If fixing the issue in the background is not possible, could you please implement a warning function in JSL such that the user gets informed about the bad performance due to non-pre-evaluated variables. Then he can adjust the code, with substitute Should you Loop through a data table or use Recode, or use Get Rows Where to change values in a data table  or Eval: https://community.jmp.com/t5/Discussions/Should-you-Loop-through-a-data-table-or-use-Recode-or-use-Get/m-p/580751/highlight/true#M78769        more wishes submitted by   ... View more

I realize that there is a workaround for doing a Friedman's test, but it is certainly not a big ask to have it in the program.   This is especially true if using it for teaching . ... View more

What inspired this wish list request?  JMP is super useful in the bioassay world. However it is not easy to constrain asymptotes especially when plotting multiple curves from the same plate. Also using applying the 4PL fit curve in graph builder is not easy. it should be available just like linear regression. This would really make your program powerful.    What is the improvement you would like to see? Please add 4P Hill fitting equation to your Graph Builder platform, just like thelinear regression and other options you have. It would just be a button icon as an option. Even more powerful would be the option to constrain the asymptotes within graph builder. The constraint option in non linear graphing is very cumbersome and often does not give the same results as GraphPad Prism which is the gold std for graphs in the biologic community. JMP is so much more powerful because of the way you can easily look at data and change axes etc, this would make your program superior to GraphPad and many more users would come to you if this was an option.   Why is this idea important? 4P Fitting is extremely important in many biological assays. If you want to extend the reach of your program making this easy will do that. We constantly have to toggle between graph pad Prism. Prism is not ideal because data import is very cumbersome so we end up having to import data with JMP and then copy paste into Prism.      ... View more

Hello,   As an alternative to the Normal Quantile Plot in Distribution, can the ability to generate other plots (e.g., lognormal, Weibull, etc.) be generated? In effect, plots should be available for all of the Continuous Fit options. This is basically bringing in functionality similar to that in Life Distribution.   Thanks and regards Mehul Shroff ... View more

What inspired this wish list request?  I saw that in JMP 17 we can now change the value order of a variable in Graph Builder, and it will save that property back to the column.   What is the improvement you would like to see?  I would like to see this same feature applied to the Tabulate platform.  I often build summary tables and this is where I frequently become aware that I need to fix the values in a particular order other than their occurrence or alphabetically.  Being able to do this right from the Tabulate page would be great.   Why is this idea important?  It's efficient and consistent with other platforms.   ... View more

I was recently running OC curves to determine AQL and RQL figures. Of interest was the difference in these figures when different sample sizes are used. It would be great to have the option of entering multiple sample sizes (along with population size, and allowable defects), and then plot the overlapped OC curves on one plot to help you visualise the ‘trade-off’ that exists in AQL/RQL figures for various sample size options. At present, JMP will only show curves for individual sample sizes. ... View more

The tolerance intervals in the distribution platform require you to adjust the proportion and confidence levels for each response individually.  It would be great if a preference could set the confidence, proportion and 1-sided/two-sided along with the option to remove the input prompt when performing the analysis. ... View more

I currently have scripts that will perform both tests, but it would be very convenient if they were built into JMP. ... View more

In an unequal slopes ANCOVA model the individual slopes can be estimated with the Estimates > Custom test or Estimates > Compare slopes option in the output for Fit model. Standard errors and p-values are also given but I miss confidence intervals for the slopes. This should also be possible if additional random effects are added to the model (a mixed model). ... View more

Add a test to produce p values of superiority tests at a range of distances and output those to a table or graph as shown below.   Consider a trial where a process change is made.  Results are compared with current performance and a t test determines that the process was improved (the mean shifted) with a p-value of 0.05 and a superiority test in JMP 17 indicates that the mean shifted by at least 0.01 with a p value of 0.12, but neither tells a whole story.  To make an informed decision, we would want to know information like: 5% chance the process stayed the same or got worse 90% chance the process improved by at least 0.01 50% chance the process improved by at least 0.006 5% change the process improved by at least 0.011 Here is a post with two potential methods of calculating these values: Compare means test: graph probability by difference in means  ... View more

I have one request/recommendation for a future release of JMP.  In the one-sample equivalence test platform in JMP 16.0.0, I noticed that there is no correction for the non-parametric test case.  Yes I can, in essence conduct the test using the one-sample T-test platform: and by running the Test mean option 2X and picking off the appropriate tests as shown in the graphic below:   But I can’t get the Equivalence region nor can I get the TOST framework formalized in this same way which is what I like about the Test Equivalence Platform for the normality assumed case. So formally the request is for a non-parametric version of the TOST for the one-sample test.    Note that for the 2-sample case I can compute a non-parametric confidence interval on the median difference and compare that to the practical difference region and effectively run the TOST in this way (e.g. Hodges-Lehmann confidence interval on median difference using Wilcoxon all-pairs NP test). This is just the “confidence interval version” of the TOST.  So this actually works quite well for the 2-sample case without any need for an additional feature in the Equivalence Test platform in Fit Y by X.   But in the 1-sample case, in the Distribution platform, I cannot do the same, why? The only way I know of doing something similar to this would be using Custom Quantiles in  JMP's Distribution platform, and that does provide a non-parametric estimate (albeit not the Hodges-Lehmann one, and I have trouble specifying the exact desired coverage for Confidence level of the TOST which is a standard 90% for a 5% significance level).   So (lower-priority) an additional feature altogether could be, the calculation of a non-parametric confidence interval estimate using Hodges-Lehmann on the median in the Distribution Platform. Or perhaps a boot-strapped confidence interval (which you can do for almost every table already in JMP Pro with Right Click > Bootstrap on data table output).   ... View more

Hi, I think it would be a terrific idea to enable the option to turn on sample size (n) within Least Square Means Tables within the Fit Model Platform, particularly for cases of retrospective data analysis where sample sizes are often unbalanced across groups.  The reason being is that the platform frustratingly shows you the total sample size for the model, but doesn't show you how many samples belong to each respective variable you are testing.     Also, while the LS Means Plots are an excellent visualization tool for comparing the LS means by group, they can be somewhat misleading to the statistically naive because they don't illustrate where the individual data values are located in relation to the mean location and LS mean error bars.  For this I would recommend allowing the user to turn on the points as an overlay to the current Least Square Means Plot.     ... View more

Extend Fit Ordinal Logistic by the following two features: - Add the Odds Ratios together with 95%-CIs, - Add a test to check the proportional odds assumption. ... View more

What inspired this wish list request? I was using the simulator platform in the Profiler to run a simulation experiment and I needed to change a bunch of the factors to normal truncated.  This took a lot longer than it should have as I had to change each manually.   What is the improvement you would like to see? I would like to see the ability to set the distribution types and the Distibution parameters from a reference data table.     Why is this idea important? For more complicated column types in the Simulate model or for trying to experiment quickly with different process factor behaviors this will greatly increase user speed and platform ease.     ... View more