JMP Wish List

Anonymous · ‎01-26-2018

Please bring the iOS graph builder app to the iPhone! Being limited to iPad makes it much less useful than it could be.

AlexanderKrebs · ‎01-24-2018

I often use the Set Title() message to overwrite the OutlineBox titles of charts, distributions, etc. Some of these titles are very long. And I would like to split these titles into multiple lines. I propose to add a Set Wrap() method to the OutlineBox. See also the related Community Discussion: JSL How do I wrap the title text of a OutlineBox?

AlexanderKrebs · ‎01-24-2018

Variability charts allow to optionally plot Grand Mean and Grand Median lines. While Grand Mean automatically adds a label of the overall mean to the right of the plot, Grand Median does not show any label. My proposal is to allow for showing a label for the Grand Median the same way as for Grand Mean. I propose that the look and feel for Grand Mean and Median shall be the same: Show a label by default, allow the user to turn on/off the label in the graph context->Customize...->Grand Mean/Median properties.

cwillden · ‎01-22-2018

This is probably the most frequent limitation of JMP at my organization. Unequal variances suggest to use Welch's ANOVA, but options are limited for multiple comparisons with unequal variances. Steel-Dwass doesn't work well with small sample sizes and it's not clear that it is robust to very unequal variances. Games-Howell test is effectively a Welch's version of Tukey's HSD and works well with small sample sizes.

cwillden · ‎01-22-2018

In Control Chart Builder, you cannot save control limits if subgroup sizes are not constant. Of course, this is because control limits are a function of subgroup size (in addition to the mean and variance). However, this makes it difficult to lock down control limits for Phase II. With fixed values for the mean and variance, control limits could still accommodate variable subgroup sizes in Phase II charting. The column property "Sigma" already exists, but we would need an additional column property for the mean. Proposal: Add a menu option to Control Chart Builder "Save Average and Standard Deviation" or something similar. Result is column properties "Sigma" and "Average" are populated. These are used to construct control limits when the column is plotted in CCB and is indifferent to subgroup size.

john_madden · ‎01-19-2018

Autosaving data tables should have an option to rename and retain the existing file as "table.jmp.bak" before saving the new version as "table.jmp". The reason is to prevent autosave unintentionally overwriting a "good" copy of a table with a (mis-)edited version. An even better option is be able to designate a backup folder where N previous versions of the file can be optionally retained.

john_madden · ‎01-19-2018

Analyze > Screening > Explore Outliers, would like an option to output the results as an indicator column ((0|1), where 1 indicates outlier. Currently, the result options are: Select Rows, Color Cells, Exclude Rows, Color Rows, Add to Missing Value Codes, Change to Missing. Of course, it is pretty easy to just duplicate the data column, Change to Missing, and then by script or formula turn that into an indicator column. But it would be nice to have Make Indicator Column as a native option.

bfoulkes · ‎01-17-2018

When I run the Life Distribution platform and create a model, there is an option to do a custom estimation. It would be good to have an opportunity to do a conditional probability calculation, in that the item has already run to time X and what is the likelihood of surviving to time Y? This would require one more input, maybe called "starting point" or something along those lines.

bfoulkes · ‎01-17-2018

I will occasionally have a distribution with zero's in there and when I try using the Distribution platform, I get a much smaller list of possible distributions to fit than normally. I then can go into Life Distribution and look at a ZI Lognormal or ZI Weibull, but why not offer these in the Distribution platform?

john_madden · ‎01-17-2018

Would like a new column property (analagous to Range Check and List Check) for character columns. It would (interactively) validate column input against a pattern specified as a regex or as a search string, and reject input that failed to match the desired pattern. My purpose is to make it less error-prone when manually filling out tables with data like id's, names, etc. that are supposed to conform to a know pattern.

scott1588 · ‎01-11-2018

As a long time JMP user, I would like to make two suggestions for new features. 1. Ability to color code column headers. This would make it tremendously easier both to organize datasets with many columns and also to more easily find columns. This could be integrated with the grouping column function to color code a group of columns. This color coding could also extend to the column list on the side of the table. 2. Second level header. When grouping columns, it would be wonderful to see another header level above the existing levels spanning the group with the group name and color. Again, this would help in column navigation and table organization. Thanks for your consideration.

yvesprairie · ‎01-10-2018

Are there any plans to integrate GAMs (Generalized Additive Models) to the predictive modeling platform. Since the procedure already exists in SAS, one would think it might be reasonably easy to port it to JMP. Many thanks, Yves P.

rocker · ‎01-09-2018

As computing has advanced, I've gone from 1 core to 2 cores to 4 cores... to 64 threads on my workstation. Meanwhile .jsl has effectively remained single threaded. I know some platforms will utilize all threads, but I'd like to be able to farm out simple jsl operations at will to multiple threads. For example, if I have a script which operates row-wise on 10,000,000 rows of data, if I'm iterating by row in a for loop and I know each row's result is independent of the other rows, I'd like to split that operation into 64 asynchronous for loops, pause other execution while the 64 loops complete in parallel, then move back to a single threaded, synchronous script after all are complete, much like I would in .py with Pool() and poolObject.join.

walter_isaac · ‎01-04-2018

See the attached journal. This is a simple example of a variability/gauge plot. from one of the sample datasets. I do a lot of beginning JMP user training, and I always show them the variability chart because it is so useful for visulaizing the effects of different X factors on a Y response all on one plot. I would like to be able to do more statistical analysis off this platform. I am looking for the same analysis that is on the fit Y by X platform. For instance in this case an Anova/Tukey means comparison by Gender. And/or a T test comparing gender by each drug.

ih · ‎01-04-2018

Two improvements to linked tables: Provide a method to automatically open linked tables when a data table opens. Include some way to disable it (for large or missing linked tables). Remember which referenced columns are visible and unhide them automatically if the linked table is available. Inspired by this post.

stephen_pearson · ‎01-03-2018

It is common to have a recently opened files and be able to pin items on this list. Why not extend this idea to the JMP menu/toolbar items? How about a sidebar (with oversize icons?) where 'favourite' menu/toolbar items appear and can be pinned. The number of items that appear on this list could also be customised. I could create a favourites menu/toolbar manually but that would require keeping a mental note of which menu/toolbar items I use most. Being able to expose the usage statistics to the end user might also prove enlightening?

tim_reeves · ‎12-29-2017

I would like the ability to change the y axis type to Logarithmic on control charts.

jwiltsie · ‎12-21-2017

Random effects should (almost) always have a 'Nominal' modeling type. I have seen multiple times where JMP users have random effects in their model with a 'Continuous' modeling type and the results are incorrect and the issue is difficult to spot. If JMP gave a warning message if any random effects were not 'Nominal', it would greatly reduce the problem.

DIVIDESbyZER0 · ‎12-21-2017

It would be great is the JMP report for a control chart could display the run rule violations that are being tested (Not just the ones that have been violated). Ideally this would be some sort of footnote underneath each plot. This would help insure that the correct run rules were evaluated during a review.

thomasz · ‎12-20-2017

Instead of the dreaded ###, please show which file and line number a script failed in. Kind regards Thomas

JMP Wish List

Graph builder for the iPhone

Add Set Wrap() Method to OutlineBox

Add Label for Grand Median in Variability Graph

Games-Howell Test (Tukey HSD with Welch's Correction for Unequal Variances) in Fit Y by X

Control Chart Builder: "Save Average & Sigma" as alternative to "Save Limits"

Option to retain backup on autosave

Indicator Column option in Explore Outliers

Add Conditional Probability to Life Distribution

Add Zero Inflated distributions to Distribution Platform

Pattern Check column property

Suggestions for new features to aid in better table organization

Generalized Additive Models (GAMs)

Multithreading in JSL

Variability Chart enhancements

Automatically open linked tables and remember referenced column hidden states

Dynamic List of Frequently Used Menu/Toolbar Items

Logarithmic y-axis for control charts

Give a warning message when random effects don't have 'Nominal' modeling type

Add Footnote that contains Control chart tests or run rules that have been applied

Line numbers on errors in log output