JMP Wish List

ttreynor · ‎05-09-2018

I find that when I am viewing my model with both the Profiler and the Contour Profiler that I usually choose option to Link Profilers. I've run Get Platform Preferences() in the Log, but it seems there is no option to set "Link Profilers" as preferred/default behavior. If there is a reason that users are not allowed to set this preference either via GUI or via JSL, what is it?

fabones · ‎05-09-2018

I would like to see the ability to fit Cox models with time-dependent covariates, as these are very common in retrospective, medical research. While the creation of (start,stop) counting process data sets might be challenging in JMP's point and click environment, perhaps an approach of submitting "wide" data sets to JMP's Cox platform with variables storing the time change-points and the corresponding new values might work (as in Paul Allison's excellent book "Survival Analysis Using SAS").

fabones · ‎05-09-2018

I'd like to see a propensity score platform in JMP with similar functionality to SAS proc psmatch, namely fitting the propensity score model, propensity score assessments, and matching. Being able to incorporate the propensity scores in other JMP model fitting platforms (e.g. stratification and robust variance estimators) would also be a big plus.

gerd · ‎05-07-2018

It would be nice if Tolerance Intervals calculated in the Distribution Platform came with a visualization (e.g.) on top of the Histogram. The graph below gives an example. There, I added the Graph Builder visualization of a Tolerance Interval below the Distribution Histogram. I left out Spec-Limits there, but if so specified in the column properties, they should be shown there, as well. Similar to other features in Distribution I image it would be possible to show or hide this visualization via the red triangle menu. There are further ideas presented in the Wish List in connection with Tolerance Intervals, e.g. https://community.jmp.com/t5/JMP-Wish-List/Setting-Confidence-and-Proportion-to-Cover-for-Tolerance/idi-p/51188 https://community.jmp.com/t5/JMP-Wish-List/3-Sigma-Intervals-and-Tolerance-Intervals-in-Graph-Builder/idi-p/51192 https://community.jmp.com/t5/JMP-Wish-List/Statistical-rationales-for-sample-sizes/idi-p/52052

kevin_lennon · ‎05-04-2018

I would like for JMP to be able to open Apache Parquet files and handle them in the same way that HDF5, JSON etc...can be handled. Thanks a million.

jlrouquette · ‎05-03-2018

I would like to see the RSquare validation value available for each row in the output table with the all possible models tool. I find myself using the all possible models option often but I always end up needing to manaually create the exact table to include the validation information. The information is already avaialbe without actually running the model in the stepwise regression control on the top of the page.

Reinaldo · ‎05-03-2018

Hi JMP users, I used to run some non-parametric comparisons for each pair using the Wilcoxon method in a repeated-measures designs through Analysis -> Fit Y by X report. I would like to know if it is possible to develop an effect-size calculator (add-in) for that purpose, please. Thank you.

Reinaldo · ‎05-03-2018

Hi JMP users, I would like to know if it is possible to develop an add-in to calculate the effect-size for repeated-measures design with random effect. I used to run the LSMeans Tukey HSD algorithm in Fit Model report. Thank you.

MathStatChem · ‎05-01-2018

See this post: https://community.jmp.com/t5/JSL-Cookbook/Combined-and-Connected-Number-Edit-Box-and-Slider-Box/ta-p/54255 It would be nice for the Slider Box() object to have the option to automatically have a connected Number Edit Box, and have both the Number Edit Box and Slider Box execute the same JSL when the either the slider is changed or the number is edited in the Number Edit Box.

msharp · ‎05-01-2018

My coworkers were making fun of me for coding with bright backgrounds. I could use another editor like Sublime or Atom, but JSL Text Editor does come with a lot of convient functionality. Anyway we could include additional themes? Preferably darker ones that are easier on the eyes.

mikedriscoll · ‎04-26-2018

Someone already requested a better search and replace function in the JSL editor. I agree with that because the existing one seems confusing to me. But what I would really like to see are bookmark capabilities in the JSL editor, with control via key strokes. Ex: ctrl+F2 toggles the bookmark on / off. F2 moves to next bookmark. Shift F2 moves to previous bookmark. Related: There's an F3 function where if you have text highlighted it will skip to the next time that string is encountered, but it would be nice to have a shift-F3 that goes in the opposite direction (up). Never mind this last part... I didn't realize ctrl+shift+F3 goes to previous. Thanks, Mike

egon_gross · ‎04-19-2018

Hello, I got JMP 14 installed on my desctop from our IT department recently and wanted to change the standard icon on the desctop by "right klick-properties-change Icon/ symbol". Unfortunately this didn't work as administrator rights are necessary to perform this procedure. As I participated in the early adopter program (and will continue to do so) it would help a lot to distinguish the program of choice right before starting the program. As the modification of the desctop icon is not such a fundamental change it would be nice to get this done even without admin rights. Yours E. Gross

d_barnett · ‎04-18-2018

When you create a model using the DoE platorms there is the option to simulate responses prior to the table being made but when you run the model script you have generated the :Y response is included, this means you have to change the model to use the :Y simulated responses you have just generated. The idea is that if you Simulate Responses during the generation of a model the ':Y Simulated' should be included into the model dialogue, not the ':Y' David

mjoner · ‎04-17-2018

Hi, I'm looking at the Profiler red triangle, under Optimization and Desirability -> Set Desirabilities. Several of our newer users get confused when they have multiple y variables because JMP pops up a separate modal box to set desirability for each response one at a time. The astute ones notice that each box is for a different response, but others don't notice the subtle change. Any way we could replace this interface with a single window? Use a List Box or a Combo Box or a Tab Box to arrange the different responses. Then the user can go to "Set Desirabilities", see all the responses in one window, and potentially set all of them from the same window. Or, with flexibility to choose which responses need to be set, having a "select the response to change" property allows the user to skip over two or three responses that don't need to be changed in the first place. Icing on the cake: allow some sort of "Standardize Desirabilities" option from the Profiler as well. Sure, you can standardize attributes from the data table and set the Response Limits property, but when you're in the middle of a Profiler that isn't intuitive.

frankderuyck · ‎04-17-2018

Many DOE's need to be peformed over several days so considering and separation of random time variation in DOE analysis is an issue. As stated by Goos & Jones Randomized blocks in DOE are very useful to absorb random variation so fixed effects can be precisely estimated. However in the JMP forward stepwise parameter selection module random effects can't be included; all blocks, also random time effect blocks, need to be treated as fixed effects. In forward selection it would be a major imporvement to also allow random effects so the old cumbersome backward selection can be avoided.

sjjanis · ‎04-11-2018

Most products have multiple quality characteristics, some of which are correlated. Multivariate process capability analysis incorporates those correlations to penalize the process when the process produces output that doesn't meet the typical relationship between the characteristics.

egon_gross · ‎04-09-2018

Hi, after the implementation of the Functional Data Explorer it would be very helpful to extend the transformation possibilities in order to handle spectroscopic (e.g. NIR/ MIR-data) inside this nice and handy platform. To use this kind of data I use "UnscramblerR" on a regular base to convert reflectance data to transmission data, change wafelength to wavenumber, do Kubelka-Munk and/or S. Golay transformations, calculate derivatives, smoothen spectra curves and so on. But as it is quite work intens to import the data into Excel, go to Unscrambler do the calcuations - sometimes more kinds than realy necessary (which I detect afterwards using JMPPRO), back to Excel and than finally to JMPPRO to dig inside the data, it would be a big releave to do all this things just inside JMPPRO. Kind Regards E. Gross

scottahindle · ‎04-04-2018

Why not simplify naming in JMP and call the Cp and Cpk values as "capability" and the Pp and Ppk values as "performance", i.e. drop the term "overall sigma capability"? Since Pp and Ppk are estimates of past performance based on the descriptive statistic "overall SD" it seems to me misleading to label them as "capability". I think it would be helpful to have capability more clearly aligned to stable/ predictable processes i.e. "within sigma" and leave the Pp/Ppk values as simple (and often useful) descriptions of past performance.

shannontyp_77 · ‎04-04-2018

Entering a random effect into Generalized Linear models generates a warning message saying that "Random effects are not supported in Generalized Linear Models". Please add support for random effects in Generalized Linear Regressions so we can enter individual identification as a random effect to avoid the effects of pseudo-replication.

Jmpkom · ‎03-29-2018

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.

JMP Wish List

how can I set preference to link profilers

time-dependent covariates in Cox models

propensity score platform

Graphical Representation of Tolerance Intervals in Distribution Platform

Opening Parquet Files

Adding an RSquare Validation Column to the All Possible Models Table

Effect-size calculator (add-in) for nonparametric comparisons (Wilcoxon) repeated-measures designs

Effect-size calculator (add-in) for repeated-measures design with random effect

Combined and Connected Number Edit Box and Slider Box

JSL Text Editor Color Themes

Add bookmarking to JSL script editor

Enable easy modification of desktop icon without admin rights

:Y Simulated Responses values used in Model Dialogue not :Y

Profiler Set Desirabilities

Include random effects in forward stepwise parameter selection

Multivariate Process Capability

Kubelka-Munk, S.Golay, reflectance to transmittance, ...

Start labelling "overall sigma capability" as "performance indexes" in JMP

Random effects are not supported in Generalized Linear Models

Overlay multiple OC Curves onto the same chart