JMP Wish List

What inspired this wish list request? I am using FDE to create FPC Scores to use as inputs to into a Generalized Regression model.  I would like to use this model outside of JMP Pro as part of a quality Control Process.  While I can use the GenReg model outside of JMP, I have no way to generate FPC Scores from the time series to use as input.     What is the improvement you would like to see? I would like to be able to to export code that takes the time series as input and exports FPC Scores.  If that is too difficult, at least create a way to generate FPC scores from new data inside of a JMP sheet.  Ideally both solutions.     Why is this idea important? This idea is important because while creating a solution in JMP is great, being able to use the solution in a quality control process is necessary.     ... View more

I was analyzing a bunch of historical data for engineers who had several potential multiple response column factors. They have only regular JMP (not JMP Pro), so several steps were needed to make them a Profiler they could play with in their copy of regular JMP.   JMP17 and JMP18EA8: I discovered that I could not get a profiler to appear in GenReg with the multiple response column factor in the model, much less if I saved the prediction formula (and PredSE formula) and used Graph>Profiler. However, if I edited the prediction formula column to remove ContainsItem() and replaces with If(Contains(),1,0) and replace the multiple response column reference with a nominal column, then Graph>Profiler would work fine. And I could give my engineers a Profiler they could play with. But these are a lot of annoying extra steps. Save me the steps!      ... View more

What inspired this wish list request?  Presently, the Stability feature only addresses FIXED BATCHES of a stability product only.   What is the improvement you would like to see?  This will allow re-estimation of a shelf-life when additional stability data of more batches of a product are generated and having the shelf-life based on a random number of batches.   If I am contacted, I can forward some articles recommending the model of random intercepts and random slopes. Why is this idea important?  Once more data are accessible, a more realistic shelf-life can be determined for all future batches produced.   ... View more

What inspired this wish list request? A highly important function the GLIMMIX procedure that I terribly miss in JMP.      What is the improvement you would like to see?  Please include an equivalent to the GLIMMIX option: Random _resid_ / group=Treatment;   Why is this idea important?  In life science heterogeneus variances are the rule and not the exception. Therefore it would be extremely helpful to be able assign different variances to each experimental group.     ... View more

What inspired this wish list request? Please describe the current issue that needs improvement or the problem to be solved that is not easy or possible right now, with an example use case.  I recently ran into some reviewers requesting that we follow the methodologies described in: Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) described in the Annals of Internal Medicine 2015; 162: W1-W73 paper. In that paper the authors described depiction of predicted outcomes (x axis) vs actual outcomes (y axis) using a calibration plot. They go on to describe net benefit curves (A Simple, Step by Step Guide to Interpreting Decision Curve Analysis; Vickers, van Calster, and Steyerberg; Diagnostic and Prognostic Research, 2019; https://doi.org/10.1186/s41512-019-0064-7) as a tool to assess benefit vs threshold risk   A tutorial and code for this tool is available online for Python, R, STATA, and SAS: Decision Curve Analysis Tutorial (danieldsjoberg.com)   This tool extends the utility of existing tools used to assess the accuracy of predictive models such as AUC (c-statistic) by allowing for the application of clinical utility. It is a tool that is increasingly used in business and other risk-benefit settings. Although output can be tabular, this this is primarily a graphic-based tool that 'fits' nicely within the JMP paradigm.   Current JMP version: JMP Pro 17.2   ... View more

I am analyzing waveforms output during RF test.  Waveform overlays of multiple board tests show board to board shifts in phase and amplitude.  I believe a useful analysis option could be using a reference/target curve as a starting point and then having the ability to calculate an envelope around the curve.  Perhaps we could say this would be a LL and UL for each target curve data point.  The calculation for the envelope curve points could have some selectable options such as +/- n sigma, or a +/- a percentage of the data point.  With the envelope drawn, compute some statistics such as N waveforms exceeding either the LL or UL, N points exceeding either the LL or UL, and an indication and list of the out of spec waveforms by some selectable parameter such as serial number.   My difficulty is setting limits at the board level is complicated by boards which can be tweaked back in to a precise operating point during system calibration.  Therefore, if we can confine acceptable waveforms to be within a specific window, we will also know there is headroom at system calibration to tweak the boards in precisely.   JMP 17.2.0   ... View more

Would like to be able to generate random values from a Weibull distribution where the threshold parameter is not 0. ... View more

What inspired this wish list request?  2つの異なる母集団のデータの同等性を統計的に記述したい。頻度論的な統計学では同等性の検定の限界があるのでベイズ統計を使用したい。一般的なベイズ統計は複雑な積分を解く必要があるためMCMC法を搭載したアルゴニズムを搭載する必要があるが一部の手法は簡潔化されており複雑な積分を解く必要はない。下記に簡潔化したベイズT検定の例を示す Rouder’sMethod G¨onen’sMethod Hyper-PriorMethod   以下文献です [G¨onenetal.,2005]G¨onen,M.,Johnson,W.O.,Lu,Y.,andWestfall,P.H.(2005).Thebayesiantwosamplettest.TheAmericanStatistician,59(3):252–257. [Kruschke,2013]Kruschke,J.K.(2013).Bayesianestimationsupersedesthettest.JournalofExperimental Psychology:General,142(2):573. [Patil,1965]Patil,V.(1965).Approximationtothebehrens-fisherdistributions.Biometrika,52(1/2):267271. [Rouderetal.,2009]Rouder,J.N.,Speckman,P.L.,Sun,D.,Morey,R.D.,andIverson,G.(2009).Bayesian ttestsforacceptingandrejectingthenullhypothesis.Psychonomicbulletin&review,16(2):225–237. [WangandLiu,2016]Wang,M.andLiu,G.(2016).Asimpletwo-samplebayesiant-testforhypothesis testing.TheAmericanStatistician,70(2):195–201.   What is the improvement you would like to see?      Why is this idea important?      ... View more

What inspired this wish list request?  I am trying to fit a pharmaceutical stability analysis with Fit Mixed (as discussed, e.g., in Statistics for Biotechnology Process Development). However, I cannot replicate the corresponding SAS analysis because the approved methodology uses the SAS default for DDFM (CONTAIN). The current JMP hardcoded variation of Kenward Roger (which appears to be accurate and similar to analogous PROC MIXED output) gives me volatile DF estimates that render the current implementation in JMP unusable. I understand the benefits of KR, but this is a set procedure (sort of like the corresponding fixed effect stability analysis in the Degradation platform) that I need to match.     What is the improvement you would like to see?  Add at least one of the other DDFM options similar to PROC MIXED for Fit Mixed and/or Fit Least Squares.     Why is this idea important?  This is a common application for pharmaceutical customers and they may need to use SAS instead.     ... View more

hi,   do you think you can add the option to choose different loss functions when modelling? Huber loss function is very effective handling outliers when using decision trees ... View more

  What inspired this wish list request? JMP does not seem to have a built-in support for zero-inflated models?    What is the improvement you would like to see? Possibility to fit zero-inflated model in JMP   Why is this idea important? It occurs quite often that the response is zero-inflated. Zero-inflated models are used when the response variable has a large number of zero values and a continuous, non-negative distribution for the non-zero values. Zero-inflated models combine two separate models: one for predicting the probability of a zero outcome, and another for predicting the non-zero values.     ... View more

What inspired this wish list request?    I have some data that i'm trying to fit a non-linear formula to (in this case a power law), but i have variants generated at different process conditions, so there are multiple variables (in this case 2). Ideally i'd like to be able to generate an equation capable of predicting the outcome (y variable) as a function of the x variable and another 'process setting' variable. The graphs look like this when performing the non-linear regression, using the 'grouping' function:   I'd love if the software were capable of 'parameterising' the 'legend' variable - ie to be able to predict the y-value based on the current x-value and the current legend value. At the moment it just groups them, but this isn't as useful as it could be...   What is the improvement you would like to see?    I'd love if the software were capable of 'parameterising' the 'legend' variable - ie to be able to predict the y-value based on the current x-value and the current legend value. At the moment it just groups them, but this isn't as useful as it could be...   Why is this idea important?    In engineering where i work there are numerous times where i'm faced with non-linear trends and data, where i don't know the underlying physics or the equations of the curves. It would be a huge advantage to be able to perform a 'multiple non-linear regression' on such data to be able to extract a predictive or descriptive capability.    ... View more

What inspired this wish list request?    A few of us have been pitching work-arounds for this here: https://community.jmp.com/t5/Discussions/multivariate-data-with-a-repeated-measures-design/m-p/632672?utm_campaign=cm&emcs_t=S2h8ZW1haWx8dG9waWNfc3Vic2NyaXB0aW9ufExIUzMzVVdSU0haTkY3fDYzMjY3MnxTVUJTQ1JJUFRJT05TfGhL&utm_medium=notification&utm_source=general.  I wonder if this implementation is not as daunting as it would seem. Could features of the mixed-model platform, namely specifying the repeated subject, be incorporated into the partial least squares platform?   What is the improvement you would like to see?      Why is this idea important?      ... View more

What inspired this wish list request?  StepWise Fit incorporates statistically invalid correlations into the model fit when there are a large number of parameters which are fit. JMP Stepwise criteria is only looking at Sum of Squares along with the F Ratio to produce what stays in the model.  Then only when you Make and Run the final Stepwise Model do you get the Fit Group Effect Summary that gives you the option to get FDR Logworth and FDR PValues. An FDR alpha should be the default input for StepWise rather than the current inputs.    We had to research why JMP is failing to select the correct criteria for incorporating parameters into the fit. Project Euclid (https://projecteuclid.orgtalking)  talks about incorporating FDR in Stepwise and we implemented into Python since JMP could not perform this properly.      What is the improvement you would like to see?      Why is this idea important?      ... View more

What inspired this wish list request?      What is the improvement you would like to see?  Here is the SAS code:   proc phreg; by cell; class code/param=glm; strata package set; model &v=code/ties=discrete; lsmeans code/pdiff lines; run;quit;   Looking to be able to do this analysis in JMP.   Why is this idea important?      ... View more

I perform multivariate linear regressions using the stepwise statistical analysis method, and when I need to remove variables with a high VIF in the coefficient estimation section, I have to search for them among the summary of effects section to select and remove them. The improvement I would like to see is the ability to directly select variables in the coefficient estimation section, which would automatically select the corresponding variable in the summary of effects section for removal. This would save a lot of time when conducting this type of test with a large number of variables.   ... View more

What inspired this wish list request?  Data mining "bump hunting" algorithm that does not require an empirical model.  Rather, it 'patiently' / incrementally searches hyper-boxes of input factors that maximize the response.   What is the improvement you would like to see?  From R: library(prim) library(MASS) data(Boston) x \<- Boston\[,5:6\] y \<- Boston\[,1\] boston.prim \<- prim.box(x=x, y=y, threshold.type=1)   Why is this idea important?      ... View more

What inspired this wish list request?  I'm using JMP Pro 17 and really happy with the introduction of Design Space Profiler. However, I was surprised to see that inside some models' platforms that can use Profiler, the Design Space Profiler was not directly accessible/available (example here with SVM, not having the possibility to directly use Design Space Profiler feature, but this "bug" is also present for Bootstrap Forest, Boosted Tree, and perhaps some other modeling platforms) :  For some other models' platforms, the Design Space Profiler is directly accessible/available (here with Neural Networks) :    What is the improvement you would like to see?  There is still a workaround for this situation, by saving the formulas in the datatable and create a Profiler based on these formulas. The Design Space Profiler is then always available :  But it would be easier if the Design Space Profiler was available "by default" in all modeling platforms that have access to the Profiler.   Why is this idea important? The Design Space Profiler is a very informative and interesting feature. JMP beginners may not be aware of the possible workaround, and may miss some very interesting informations because of the "non-native" presence of this feature in all modeling platforms. ... View more

What is the improvement you would like to see? Add a local data filter option to Non-Linear Specialized Modeling.   What inspired this wish list request?  I have to use a non-linear model to record differences between trials, so it is sad that I can't filter like other features in JMP. Instead I have to do one at a time, or make a new data set from selected data from the graph builder data filter. Very time consuming for me and round about.   Why is this idea important? It will save a lot of time in rebuilding the model if I can have a data filter once the model is created. I would just have my model, sort, and hit [Go] again to reapply my model. It would be very nice for large sorted data sets. ... View more