JMP Wish List

Submitting this on behalf of Doug Sommerville, a JMP user at the US Army, .  Doug has long used Probit analysis in his toxicology studies.  JMP defaults to Logit link functions for most of this type of analysis.  The Probit Link is offered in the GLM personality of the Fit Model platform.  However, it only supports binary responses.  Doug has a need to analyze responses with more than two ordinal levels.  Allowing JMP to provide an ordered Probit analysis would greatly facilitate his work and that of anyone doing toxicology.  It is common practice in this field to compare different toxic materials by their Probit slopes.  It would also be great if this link function were offered in the Ordinal Response Personality as well as in GenReg.  This request has come out of recent Tech Support Case Number 00025561 which contains an example data set.  I have also attached the data set below. Note.  Doug is a more than 20-year user of a competitor which has offered Probit analysis that JMP did not.  It would be great to give him one more reason to use JMP. ... View more

This is simply a labeling issue, but when teaching my biostatistics course I have to constantly deal with it.  In experimental design, there is a difference between matched data and paired data.  Matched data will be two different populations in which an attempt has been made to reduce the variables by matching for certain characteristics that might impact the response being studied but which aren't the focus of the study.  For example, age, gender, smoking status, etc.  Paired data is two populations of numbers in which the same variable has been measured on the same population usually at two different times, or under two different conditions.  For example, before and after treatment with a given drug.  Matched data is properly analyzed with a standard t test, whereas paired data with a paired t test.  Now the problem is, the paired t test is found under Analyze > Specialized Modeling > MATCHED Pairs.  This nomenclature can be confusing to students particularly when they are already having problems understanding the material (and don't particularly want to be in the class anyway, but it is required).  Could this be changed to PAIRED DATA, or something other than MATCHED PAIRS?  Thanks. ... View more

Good day: The Variance Components option within the Variability Gauge plot assumes all factors are random.  Often, it  is not the case that all factors are random; in this mixed model scenario, it would be big step forward for this platform if some factors could be designated as "fixed" so that only the random effects would have variance components associated with them. Regards, Mark   ... View more

In our work, we use a fixed lambda value for our analysis. Hence, our preferred platform to use for spline fitting is Bivariate platform.   Fit Model has nice functionalities such as simulation profiler, data interpolation with multiple factors, etc. We also want to use fit model smoothing spline function, but it's hard to bridge with Bivariate because Bivariate Fit Spline takes lambda factor as input, while Fit Model Smoothing Spline takes number of knots as input.   It would be nice to have option to set lambda instead of number of knots in Fit Model Smoothing Spline, so it can easily be bridged with Bivariate Fit Spline. Example, when I want to extend my model to 2-3 factors from my current 1-factor model in Bivariate Fit Spline.   Thanks. ... View more

JMP's Bivariate Fit Spline has nice capabilities and customization but lacks option for addition of confidence intervals. It would be good to have a Bayesian Confidence Interval visualization whose equation we can also export as column formulas. We can have option to set the confidence interval to 95%, 90%, etc. ... View more

Graph Builder has hardcoded standardization X for its smoothing plots. It would be nice to have an option to turn this on or off, similar with Bivariate charts. ... View more

I know there is an add-in for the "Fit Model" platform, but it's not always the easiest to use. It would be quite easy just to add Cohen's d to the output of "Fit Y by X," and this would be quite useful. ... View more

The Power and Sample Size Explorer is lacking an option for tolerance intervals. It is often desirable to get a minimum sample size for a tolerance interval based on the desired margin of error and its probability. Faulkenberry-Weeks is such an approach. Find inspiration in what Minitab does: Methods and formulas for Sample Size for Tolerance Intervals - Minitab ... View more

Hi! Currently, JMP allow to save the DTM formula, but the response is given in a array. It would be very nice if JMP propose to save the DTM formula in columns, in the same format than 'Save Document Term Matrix'. This will be very useful when a model is fitted to predict a class and you want to extend/update the database frequently in order to predict the class of the new lines. Best regards,   ... View more

  Hello, Some students have asked about calculating effect sizes with jmp.   Although there is an add-in, it would be useful to have them in the anova and regression tables in the standard releases.    In my opinion, it would help to simply use for students and researchers. Sincerely Juan Carlos   ... View more

JMP should support an improvement to the box-cox λ parameter when transforming the response in the least squares fit model, so that the reverse transformation is automatically applied to graphs. This would save messing around for hours withg the graph builder. ... View more

I think it should be possible to add custom axis breaks to the Graph Builder, in the same way it is often possible in other graphing software like Origin Lab.   At the moment, producing a good display for continuous X-axis data with X-axis grouping is quite involved. For example: if we have continuous timestamps over multiple days, but the data goes only for one hour during each day, but you still want the graph to show the course during a single day.   Previous post explaining the problem: https://community.jmp.com/t5/Discussions/How-to-insert-X-axis-break-in-graph-builder/m-p/525673 ... View more

Hi,   In JMP screening functions such as Response screening, add an input field of threshold pValue in the GUI to allow customers to set their own thresholds for finding strong correlation, instead of JMP's own default 0.05 or 0.01 value.  These values seem to be too loose to find strong correlation between response and factors.   Thanks, Shen ... View more

Good day, I would like an ANOVA equivalent for a non-parametric test. I understand that to do this there are several options in JMP, one which is the The Kruskal-Wallis test, however, is it possible to have a two/three way type of ANOVA in the case of a parametric test that we do using the Fit model  Also, when I try using the Fit Y by X option using Steel-Dwass there are no option for connecting letters. (I see a lot of people have requested for this option on JMP, is there by any way a solution yet? I would really appreciate one). ... View more

When performing certain tests in JMP, the null hypothesis is stated, but not for all.  I found several people are confused by the interpretation of the results for hypothesis tests if the null hypothesis is not stated.  Could a future version of JMP state the null hypothesis whenever there is one to consider? ... 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

 it will be great if user can have option to show cell median and connect cell median in variability chart(box plots). Reason being that when you have very skewed data, the mean is very misleading. ... View more

The  number of decimals shown in the UCL/LCL varies from chart to chart randomly.  I cannot find a way to set the number decimals manually.  @PatrickGiuliano suggested I submit this issue via the wish list. I feel it is a bug.  ... View more

Currently Fit Y by X platform where a single Y and single X with a By role does not turn on Fit Group. But in JSL, you can wrap the script in Fit Group and it works.   Normally in Fit Y by X, when you do multiple X's or multiple Y's, you will get the Fit Group (using Big Class.jmp example file): Fit Group( Oneway( Y( :height ), X( :age ), ), Oneway( Y( :weight ), X( :age ), ) );   If you run Fit Y by X for say two continuous, and a categorical in the By() role, you cannot arrange by rows: Bivariate( Y( :weight ), X( :height ), By( :age ) ); But using JSL, you can simply wrap it inside the Fit Group and it works: Fit Group( Bivariate( Y( :weight ), X( :height ), By( :age ) ) );   as does the more general call to Fit Y by X: Fit Group( Fit Y by X( Y( :weight ), X( :height ), By( :age ) ) ); Could this be allowed to be the default, or at least allow as a Platform Preference?   What is saves to the table or script window looks like this: Fit Group( Bivariate( Y( :weight ), X( :height ), Where( :age == 12 ) ), Bivariate( Y( :weight ), X( :height ), Where( :age == 13 ) ), Bivariate( Y( :weight ), X( :height ), Where( :age == 14 ) ), Bivariate( Y( :weight ), X( :height ), Where( :age == 15 ) ), Bivariate( Y( :weight ), X( :height ), Where( :age == 16 ) ), Bivariate( Y( :weight ), X( :height ), Where( :age == 17 ) ) ); ... View more