JMP Wish List

Can you please add the ability to test for over dispersion/under dispersion and then the option to make Laney P and U charts? ... View more

I am a big fan and user of the Custom DOE option in JMP. I recently use the option to plan a pro COVID vaccination advertising campaign that significantly increased rates. At the same time, I wish that the software included expected bias (at least as an option) in the design criterion. Both Doug Mongomery and I have written papers about variance plus expected bias criteria. George Box pointed out the importance of including bias in experimental planning optimization. For related references please consider our old article: Allen, T. T., & Tseng, S. H. (2011). Variance plus bias optimal response surface designs with qualitative factors applied to stem choice modeling. Quality and Reliability Engineering International, 27(8), 1199-1210. Variance plus bias optimal response surface designs with qualitative factors applied to stem choice modeling (researchgate.net) It shows how to make reasonable assumptions in general situations. Contact me at allen.515@osu.edu if you want to discuss this. ... View more

I would like to have possibility to create column with all values (might cause issues, I know) from Summary platform. This would allow me to calculate statistics for some group and still see values for specific columns. This column could also be made Multiple Response   Example with SQL in JMP: Names Default To Here(1); dt = Open("$SAMPLE_DATA/Big Class.jmp"); Query( dt, "SELECT age, avg(height), GROUP_CONCAT(name, ',') as name FROM 'Big Class' group by age" ); Would most likely require selection for separator and to determine if column should be multiple response. These could maybe be done in same way as Quantiles which have different selector below column list   ... View more

I would like to have an option to prevent print/return on last executed row. Currently I'm ending most of my scripts with Write() to work around this. In some cases this has caused my JMP to crash because I have forgotten about this and run a script which ends up in a large matrix.  ... View more

I would like to have possibility to save JMP script when ran to user defined folder (determined in Preferences). It would be nice to choose folder, possibly affect the naming a little (timestamp) and how many files should be kept saved per script.    Currently if the option Autosave before Run is checked, the script will be either overwritten or written to temp folder once. ... View more

From time to time I would like to re-order my data how it did look when I did open it. I can do this if I remember to create column "Order" when I open the table, but I would like to have Preference in JMP which would allow me to have this created automatically either as Column or possibly some table property which would allow me to create the column when needed.    Example how I currently handle it (with addin or script with shortcut): Names Default To Here(1); dt = Open("$SAMPLE_DATA/Big Class.jmp"); If(N Items(Get Data Table List()) == 0, Throw("No datatables open"); ); new_col = Current Data Table() << New Column("ORIGINAL_ROW", Numeric, Ordinal, Hide, Exclude, << Set Each Value(Row()); ); Current Data Table() << Move Selected Columns(new_col, to First); Not sure how subsetting and such should be handled, but maybe the original order could be left there and new created due to subsetting? ... View more

I would like to be able to create Multiple Response columns when splitting data tables. This would allow user to see that there are multiple values in columns that you didn't drop (currently it takes first? value based on order?).   I can fairly easily script my own custom Split platform, but I would rather use JMP's own. Below is example script with "regular" split and what I would like to have (can already be done with combine columns but without scripting is a hassle). Names Default To Here(1); dt = New Table("MEAS_DATA", Add Rows(10), Compress File When Saved(1), Set Cell Height(19), New Column("ID", Numeric, "Continuous", Format("Best", 12), Set Values([1, 1, 2, 2, 3, 3, 4, 4, 5, 5])), New Column("Process", Character, "Nominal", Set Values( {"MEASUREMENT_A", "MEASUREMENT_B", "MEASUREMENT_A", "MEASUREMENT_B", "MEASUREMENT_A", "MEASUREMENT_B", "MEASUREMENT_A", "MEASUREMENT_B", "MEASUREMENT_A", "MEASUREMENT_B"} ) ), New Column("Value", Numeric, "Continuous", Format("Best", 12), Set Values([1, 2, 1, 4, 7, 1, 2, 6, 2, 1])), New Column("Error", Character, "Multiple Response", Set Values({"0", "0", "1", "0", "2", "1", "0", "0", "0", "0"})), New Column("Status", Numeric, "Continuous", Format("Best", 12), Set Values([1, 1, 0, 1, 0, 0, 1, 1, 1, 1])), New Column("Time", Numeric, "Continuous", Format("Best", 12), Set Values([1, 2, 1, 2, 1, 2, 1, 2, 1, 2])) ); dt_split = dt << Split(Output table("Split"), Split By(:Process), Split(:Value), Group(:ID), Sort by Column Property); //suggestion... directly from split? I could also fairly easily script this //multiple response only supports character data, but doesn't really matter in my use cases (usually) dt_mult_resp = dt << Split( Output table("Multiple_response_split"), Split By(:Process), Split(:Value, :Error, :Status, :Time), Group(:ID) ); dt_mult_resp << Combine Columns(columns(:Error MEASUREMENT_A, :Error MEASUREMENT_B), Column Name("Error"), Delimiter(",")); dt_mult_resp << Delete Columns(:Error MEASUREMENT_A, :Error MEASUREMENT_B); dt_mult_resp << Combine Columns(columns(:Status MEASUREMENT_A, :Status MEASUREMENT_B), Column Name("Status"), Delimiter(",")); dt_mult_resp << Delete Columns(:Status MEASUREMENT_A, :Status MEASUREMENT_B); dt_mult_resp << Combine Columns(columns(:Time MEASUREMENT_A, :Time MEASUREMENT_B), Column Name("Error"), Delimiter(",")); dt_mult_resp << Delete Columns(:Time MEASUREMENT_A, :Time MEASUREMENT_B); I think this should most likely be new Col List Box in Split platform (and if that isnt possible then check box, but this isnt as good option in my opinion). ... View more

Once I started using block folding, I found myself doing full collapsing all the time. It will be more efficient if a hotkey can be created for it (ctrl + '-' is suggested).   ... View more

After running this script dt3 will contain a column 'a'. When using concatenate in a script this can result in unexpected behavior or the need to manually remove variables that are not used in the table. (Example: the 'SQL' table variable automatically added to tables returned from SQL Queries.) Provide an option in Concatenate() to rename duplicate table variables before concatenation if they would result in a new column.    Names default to here(1); dt1 = New Table( "Test", Add Rows(1), New Table Variable( "a", 1 ) ); dt2 = New Table( "Test", Add Rows(1), New Table Variable( "a", 2 ) ); dt3 = dt1 << Concatenate( dt2 ); ... View more

The CDOE profiler is a helpful tool in the Fit Curve DOE platform, but an improvement would be to include confidence bands around the final prediction. This is not currently an option in the CDOE Profiler red triangle. From the Curve DOE Analysis red triangle, you are only able to save the prediction formula (not the standard errors). The individual exponential model parameters have these bounds in their profilers. It would be great to have the ability to include confidence bands on the final CDOE prediction profiler as well. ... View more

Depending what is selected in the text editor the keyboard shortcut to comment a block of code (an implemented wish list item) will either insert a block comment (/* .. */) or it will prefix all lines with a single line comment (//).  Please let the user decide which to use by separating them into separate shortcuts. I, for example, would almost always prefer to prefix lines with //.     ... View more

Is it possible to add a password requirement to open a JMP file?    I understand that in JMP, you can encrypt scripts to view or run by password but not data tables.   I have tried to regenerate the data table by creating a script “copy table script”, but this this seems a bit complex to have to update the script and starting file every time. I assume that is how this would have to work.  ... View more

Currently in a Boxplot using JMP 16.1.0, if more than one result is plotted on y-axis, the boxes can be visulaized with an "offset" or "Align" (Box placement). This function is not given for "points". I would like to have this functionality to get a better representation of my results and to avoid an intermediate step, e.g. by restacking.     ... View more

The Profiler is a powerfull tool in JMP and widely used for the analysis and interpretation of Design of Experiment models. To make prediction about future runs, the uncertainity of a model has to be considered. However, JMP is currently limited to confidence intervals, which only allow to make predictions about the range where the true mean falls into. Prediction and tolerance intervals on the other side would extend this so single runs or even for a poportion of a population. Other software tools such as DesignExpert offer these options already for years and I guess many other JMP users would appreacitae such an option in one of the future JMP versions.     ... View more

To help users interpret how predictions being explored in the profiler relate to the data used to build them, provide options to show: Histograms and/or box plots in the borders of the profiler.  Histograms were already suggested by @FN  in Add histograms to the JMP profiler .  Extend the red dotted line through them. Time series (or row series) trends in the profiler borders with the red dotted line extended through them. Scatterplots in fields behind the profiler, turned on by default. The extrapolation statistic as an additional y/predicted variable, shown at the bottom, show the UCL on the axis A scatterplot matrix showing combinations of all variables - not on the drawing below Option to scale label sizes - useful when trying to visualize large profiler, make a profiler with 20x20 variables, shrink the cells to 1/2 the normal size, and still be able to identify variable names With everything turned on, the user would see this along with a scatterplot matrix, again with the current values as a red dotted line through all plots (I used blue here):     @cwillden sorry it took so long to enter this after our conversation at the last Discovery!   ... View more

There are a lot of forum posts about how to get results from reports.  This could be made easier by adding some attributes to the display tree and defining two new functions that would do most of the work for the user.  I propose this syntax:   platform = Fit Model( ... ); platform << Get Result( "BIC", fit("NormalLassoAICValidation") ); //returns [0.683248] Display Tree Attributes Define a set of result types that is standard across all platforms. This will be a big list, but if two platform both report an adjusted R2, they should both use the same result type. Add the result type to an attribute in the display tree for the column, cell, and/or anywhere else it is displayed, similar to the helpKey attribute for some platform outlineboxes. Display the result type somewhere in the platform GUI, maybe in a hover label for everywhere it is displayed Do something similar to flag outline boxes containing a fit created by the model (think '...fit with 6 components' in PLS or 'Normal Lasso with AICc Validation').  Here the attribute would be specific to the platform. Functions Create functions to get The result values The report element containing the result The report element of the fit The scriptable object of a fit Call them like this platform << Get Result("AdjustedRSquared") //Matrix or list of all Adjusted R Squared values platform << Get Result("AdjustedRSquared", fit("NIPALS_1")) //Matrix or list of all Adjusted R Squared values in fits with NIPALS_1 platform << Get Result("AdjustedRSquared", fit("NIPALS_1"), fitnumber(1)) // Matrix or list of all Adjusted R Squared values in the first fit with NIPALS_1, should be exactly one value platform << Get Fit("NIPALS_1") //returns the scriptable object platform << Get Fit("NIPALS_1","object") //same platform << Get Fit("NormalLassoAICValidation","report",fitnumber(1)) //returns the report from the first fit with NormalLassoAICValidation These functions these could be variations of wrappers around XPath: platform << XPath("//NumberColBoxItem[@ResultType='AdjustedRSquared']") platform << XPath("//OutlineBox[@Fit='NIPALS_1']") platform << XPath("//OutlineBox[@Fit='NIPALS_1']") << Get Scriptable Object fm << XPath("//OutlineBox[@fit='NIPALS_1']//NumberColBoxItem[@ResultType='AdjustedRSquared']")   ... View more

Log plots are commonly used for showing data, but currently the format options in JMP are missing an important option. We would like for the aesthetically pleasing X^Y option for a continuous variable format that can be used to label the axis. (i.e. 10^-1, 10^0, 10^1). ... View more

ODBC libraries typically provide the ability to create parameterized queries that pass the query and parameters to the database separately.  This basically eliminates the risk of SQL injection attacks, but is also very convenient from a programming standpoint.  Because JMP has ODBC queries, it should also have parameterized queries so that you can do:   Execute SQL(database, "Select * from Employees where Name like ? and Status=?", "BobList", {"Bob", "Active"} ); pyodbc's parameterized queries: https://github.com/mkleehammer/pyodbc/wiki/Getting-started#parameters   The .NET ecosystem's parameterized queries: https://docs.microsoft.com/en-us/dotnet/api/system.data.odbc.odbccommand.parameters?redirectedfrom=MSDN&view=dotnet-plat-ext-5.0#System_Data_Odbc_OdbcCommand_Parameters   ... View more

The current behavior of the filters in the query builder when opening a query or draging in a new filter is that all distinct values of the filter are queried automatically. When doing this on large data tables, that slows your work down significantly while you are waiting for the filter queries to finish. I would like the possibility to tell the query builder to only start those distinct values queries manually and when I decide to, e.g., only after I have finished tweaking around with my filters or if I want to use this special filter. ... View more