JMP Wish List

Isolation Forest has been emerging as arguably one of the most popular anomaly detectors in recent years due to its general effectiveness across different benchmarks and strong scalability. It is computationally efficient and has been proven to be very effective in anomaly detection.   Isolation forest compared to others (src) The algorithm isolates observations by randomly selecting a feature and then randomly selecting a split value between the maximum and minimum values of the selected feature. The implementation is therefore the same as partition trees and bootstrap trees with the difference that the target function should be uniform (trying to split variables with a constant Y value). Yet, if you try this in JMP, it doesn’t work. By including Isolation Forest in JMP, similar tu predictor screening, users would have access to a powerful tool for detecting anomalies in their data. This could help them identify unusual patterns or behaviors that may warrant further investigation.   This paper studies how IForest works and improves upon its few limitations (i.e., extended isolation forest) https://hal.science/hal-03537102/document   Scikit-learn documentation https://scikit-learn.org/stable/modules/outlier_detection.html     ... View more

☑ cool new feature ☐ could help many users! ☐ removes a „bug“ ☐ nice to have ☐ nobody needs it   #myTop10_2022   If a column is set to "multiple response", there are cool options available in the data filter menu: These allow the user to get much more power out of the data filter. Most of them rely on the fact that there are "multiple responses" in the column - and the user can select which combination he wants. There is also an option "none". It has the same effect as the "inverse" - but the difference is that "inverse" just inverts the current selection, but "none" is a setting which is applied in the background. With "none" fancy Data filters are possible along the idea: THIS, but not THAT: https://community.jmp.com/t5/Discussions/Conditional-Data-Filter-for-modeling-type-Multiple-Response/m-p/556062#M77004   Unfortunately, for standard columns, there is no "none": https://community.jmp.com/t5/Discussions/advanced-data-filter-arithmetic-just-available-if-the-column-is/m-p/556682   Every Jmp user who worked with the cool functionalities of multiple response columns (esp. the availability of "not") feels an incredible lack of functionality when working with standard columns.   my wish: Add "not" also for standard data filters.     more wishes submitted by   ... View more

Recommend adding the ability to create CDF plots within Graph Builder.     Add a new "Connection" type -- maybe "Extended Step", that would draw the first step from the left edge starting at 0, and from the last point to the right edge to represent a true CDF.  Then, document the approach, and consider adding these settings as some type of preset. ... View more

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? ... View more

Like in other statistical software it would be nice to be able to calculate tolerance intervals for the following non-normal distributions: Lognormal Gamma Exponential Smallest extreme value Largest extreme value Logistic Loglogistic Weibull I know there are some workarounds and normalizing via transformation ist always a possibility, but having just to press one button would be perfect.   Best regards, Christian ... View more

it would be great to have DBScan and HDBSCAN available as they are powerful and useful for a variety of instances where flat and parametric clustering fail. ... View more

For some reason the default font on most plots is set super low and is almost always impossible to read. I’ve been able to change the default setting to a larger global font, but sometimes this isn’t appropriate and I get huge text overlaid on top of  a relatively small plot. It would be nice if there was some sort of automatic scaling factor that could be applied to the text fonts depending on how big the plot windows are. ... View more

Instead of having the method comparison addin, it would be nice to have passing bablock regression as an alternative to linear regression within Fit y by x.    Advantages:   No need to further keep the method comparision addin uptodate. Also it does not work well in the germen language environment.  Possibility to compare quickly different regression methods.   ... View more

Gaussian Processes are a very powerful modeling technique in predictive analytics. https://scikit-learn.org/stable/modules/gaussian_process.html   While JMP has GP, it is only for specialized modeling (not possible to fit using training/validation/test)     ... View more

Store alternate column names or attributes, be able to switch between those names in platforms and the data table, and easily view those names while hovering over the column name or points.   While moving through exploratory, modeling, and presentation phases of work it is helpful to name columns differently, for example depending on the audience and context a column could be represented as any of the following or more.     Unit98765.Tag12345.SP Feed Flow Controller Set Point Feed Flow Controller Set Point (Unit98765.Tag12345.SP) Feed Flow Controller Set Point (m3/s) Especially while working in JMP with an audience (for example answering questions with ad-hoc analyses during a presentation) it is helpful to have as much info about columns accessible with as few clicks as possible, but these tools could speed up work for all users.   Common places this would be handy include these activities for data tables with hundreds or thousands of columns:   Search for columns when picking columns in graph builder, and verify the correct column was picked Default names in graph builder Column contribution, variable importance, and similar tables ... View more

I would like to suggest the ability to force MFI to combine files together, even if it finds differences.   Use Case: I am combining data from two instrument platforms where the data is set up identically, but the differences are enough that MFI separates the data into two tables if I have both instrument platforms. I am aware of all pertinent differences, and just want to leverage MFI for high quality, quick combination. I would like a radio button to 'force into one file' along with "Stack Similar Files" and "One Table for Each File".   While JMP does an amazing job most of the time, sometimes I know better than JMP ;) ... View more

I want to be able to show the relationships between two groups of data.  I have seen this as a Network Diagram or a Social Network plot, but JMP can't do this right now.  There is an add-in that will create a plot that is not that interactive, but I would like this as a full plot, where items can be filtered and sizes don't need to be coded, so they can be changed on the fly. The add-in shows the use case as transactions from one group to another, but this could also be the flow of data from one group to another or a more general social network connections plot. ... View more

Hi all,  working on MacOS, I am trying to convince me to switch to iOS and a new Ipad pro.  Everything I do on my MBPro is doable on Ipad Pro, EXCEPT JMP! Yes there is JMP graphViewer, but no way to implement data in a JMP table, or to convert a xls file filled on the Ipad to JMP...   It is more and more needed to have a full version of JMP on iOS! ... View more

☐ cool new feature ☐ could help many users! ☐ removes something that feels like a „bug“ ☑ nice to have ☐ nobody needs it   What inspired this wish list request?  For linked data tables (either summary tables or linked subsets),  a click on  will take me to the original data table. This is very convenient. Great that JMP is willing to share this knowledge.   Unfortunately, there is no corresponding functionality available via JSL - to get from a linked summary table or subset table to the main table.   What is the improvement you would like to see?  For reports, one can use the message << get data table()t o get a reference for the data table. For linked summary tables and subset tables, please provide a corresponding message to get a reference of the original data table.   Why is this idea important?    Just imagine how often you generate a report from a summary table - and then face the difficulty how to save the script. Actually, it's just a combination of  a) the source script for the summary table b) the source script of the report ... but it requires ~ 20 mouse clicks to save the combined script to the main table. If there is a way to get the main table - all this could be triggered by a single click on a toolbar. cool, right?   other wishes from  ... View more

☑ cool new feature ☐ could help many users! ☐ removes something that feels like a „bug“ ☐ nice to have ☐ nobody needs it   What inspired this wish list request?  In JSL there are functions like  is control key() to get the current status of modifier keys,   unfortunately,  This function does not return the current state of the control key but rather the state of the control key at the start of another event like mouse click or key press.  Checking for modifier keys (ctrl, alt, shift) in script without mouse trapping      What is the improvement you would like to see?  As  JMP does have an internal function to get the modified key state at any time but this is not exposed via JSL. Checking for modifier keys (ctrl, alt, shift) in script without mouse trapping  ... could you please provide a JSL function to expose the current status of a modifier key :)     Why is this idea important?  This will allow JSL programmers to generate Toolbars and commands in the menu on a completely new level: depending on the status of a modifier key, different actions could be triggered (!!!)   similar add-on functionality could also boost standard platforms: - A user could specify a DEFAULT action in the preferences. If a user clicks on Subset, Summary, Tabulate, Graph Builder - holding SHIFT - the DEFAULT action is executed ... without any need to manually enter the settings. - with a control key, the action could be triggered with the RECALL settings.   other wishes from  ... View more

What inspired this wish list request?  Few posts I found from community and pinning ruler could also be fairly helpful from time to time.   What is the improvement you would like to see?  Ability to pin rulers and crosshairs to graphs (with crosshairs it should be possible to hide the aiming circle). Could also be helpful to be able to add multiple of them and possibly have some additional settings such as colors (for ruler also changing increment).   Why is this idea important?  Provides additional tools to annotate your graphs   ... View more

☐ cool new feature ☐ could help many users! ☑ removes something that looks like a bug ☐ nice to have ☐ nobody needs it     What inspired this wish list request?  In GraphBuilder, there is an option to add labels "by value" directly inside the graph. Unfortunately, there is no option to adjust the orientation of the label. Therefore, labels are sometimes hard to read: How to rotate a label?    What is the improvement you would like to see?  For axis labels and annotation text, there exist options to adjust the orientation. Please implement a similar option to adjust the orientation of "labels by value."   edit: Please also add an option to control the placement: Graph Builder Bar Graph Label Offset    Why is this idea important?  Labels in Bar graphs are often not readable and via the GUI the user doesn't have a chance to fix that. Just some JSL experts can generate graphs with readable labels. [the issue got more severe in JMP 18 doe to the new way of label placement]   With the new options, any user could make the labels readable      more wishes by   ... View more

We teach many of our engineers about paired t-tests and use the Fit Y by X platform (rather than Matched Pairs) when the format of their data set is more suitable for Fit Y by X. There are 2 frustrating aspects:   There is still an option for 't test' in the analysis, which does an ordinary two-sample t-test, which ignores the Block that you incorporated. Most engineers/scientists don't realize this. If the data is unbalanced by even 1 data point (e.g. I have 20 pairs, but there is 1 missing data point or outlier that was discarded), you get an error message saying the Blocking is going to be ignored. It's a somewhat confusing message for them, and I feel that most engineers/scientists just blindly keep following the platform output. If I had fit the same model using Fit Model, this slight unbalance in the data can easily be utilized to give the correct hypothesis testing output. I'd like to see the Fit Y by X output handle the unbalanced data when the correct model is rather straightforward. ... View more

Hello,   When using the Table-processing commands, it would be helpful to provide checkboxes in the dialog to determine how to handle hidden / excluded rows or columns. In particular, this is problematic with splitting tables.   Regards Mehul Shroff ... View more