About rc_hertzy

rc_hertzy

This series is an excellent tutorial for nonlinear modeling. However, I had hoped to find one more step: how to add a custom model to the Fit Curve platform. While part 4 did show the extra features of the Nonlinear platform, Fit Curve provides excellent comparison summaries (e.g., AICc) when fitting multiple models, and in tables that are easily scripted for capture into a Word document. Is there any way to add a custom model to the Fit Curve platform?

rc_hertzy · ‎09-02-2020

// = = = // Script to concatenate a file to an existing master file Names Default To Here( 1 ); // Open the master data table // Then call current master data table "dtmaster" dtmaster = Current Data Table(); // Open file dialog to select file to be appended to master, name it "dtnew" // If an Excel file the Excel Import wizard will open dtnew = Open(); //Concatenate the new to the master file by appending it dtmaster << Concatenate( dtnew, Append to first table ); //Close new table Close( dtnew, NoSave ); // Now the master file contains the new appended info

rc_hertzy · ‎04-04-2020

Can this be done for two Y variables? I have Y1, Y1LCL, Y1UCL, Y2, Y2LCL, Y2UCL The Graph Builder script works if only one Y variable. When I added 3 lines for the corresponding Y2 values, they only show up as points. I can show both Y vars with CLs using a Group variable when Ys are not in separate columns, but I get two graphs, one above the other, whereas I want both Ys and their intervals on one graph (thus creation of Y1, etc). Here is the script using a Group: Graph Builder( Variables( X( :ID ), Y( :Predicted Dose ), Y( :Lower 95%, Position( 1 ) ), Y( :Upper 95%, Position( 1 ) ), Group Y( :BMR% ) ), Elements( Bar( X, Y( 2 ), Y( 3 ), Y( 1 ), Legend( 16 ), Bar Style( "Interval" ) ) ), SendToReport( Dispatch( {"Bar"}, "", OutlineBox, {Close( 0 )} ) ) );

rc_hertzy · ‎03-27-2020

This is a very useful script! I also need to output Excel files, so changed the file_path and openDTS[i] lines to file_path = file_dir || one_name || ".csv"; //Use the copied name for the new .csv's name file_pathE = file_dir || one_name || ".xls"; //Same root for new .xls file openDTS[i] << Save( file_path ); //Save as csv files openDTS[i] << Save( file_pathE ); // Save as Excel

rc_hertzy · ‎10-19-2019

Elegant! Needs to add a final parenthesis at end. )

rc_hertzy · ‎10-17-2019

Worked very smoothly (JMP Pro 14.2 on Mac) except graphs are incomplete when compared against the corresponding journaled reports. The platform reports are from Fit Curve, where 3 models are used, with different reports for the different data sets modeled. The Model Comparison table at the top shows fine, along with a color for each model. The next item is Plot. For some reports, the data and curves show up fine, while others only show the data. I tried closing each Plot before running the HTML5 add-in, but that did not help. Neither did closing and opening the Plot. The same issue occurs with the Inverse Prediction plot: it shows the predicted doses and confidence intervals, but not the model curve. I have tried opening the index.shtml file both in Safari and in Firefox, same issue. Any ideas?

rc_hertzy · ‎03-12-2019

@rc_hertzy wrote: Nice solution. First, a typo (equal sign not minus sign): fourth line should be dt=data table(1); Now, how can I keep one column for the source? If I merely add Create source column in the Concatenate statement, I get successive source columns, and none is what I want - listing all source files. Related: when I manually concatenated files and copied the Source script from the output file to a script window, it would not run, giving error of not finding the listed filenames. Here is the script ( File1.jmp is the selected open file): dt = Current Data Table(); dt << Concatenate ( Data Table ( "File2.jmp" ), Data Table ( "File3.jmp" ), Output Table( "Combinedtest.jmp" ), Create source column ); Thanks. -Rick I found my error in the last part. I had renamed one data file so of course the script could not find it. Still I would like a way to modify your solution to add a source column so I do not need to explicitly list all the open files. Thanks.

rc_hertzy · ‎03-12-2019

Nice solution. First, a typo (equal sign not minus sign): fourth line should be dt=data table(1); Now, how can I keep one column for the source? If I merely add Create source column in the Concatenate statement, I get successive source columns, and none is what I want - listing all source files. Related: when I manually concatenated files and copied the Source script from the output file to a script window, it would not run, giving error of not finding the listed filenames. Here is the script ( File1.jmp is the selected open file): dt = Current Data Table(); dt << Concatenate ( Data Table ( "File2.jmp" ), Data Table ( "File3.jmp" ), Output Table( "Combinedtest.jmp" ), Create source column ); Thanks. -Rick

rc_hertzy · ‎02-26-2019

I'm using Fit Curve with 3 models but the resulting plot shows all curves as solid lines of blue or green tint; those do not distinguish well in black and white journal articles. Is there a way to customize the plot by scripting each model? Saving the results as a script gives: Fit Curve ( Y( :Liver Wt ), X( :Dose ), Fit Logistic 3P, ... While I can save script phrases for adding items (e.g.,vertical lines) via SendToReport() I don't see how to change the model curves' styles. Related: can Fit Curve be scripted to save the prediction formulas?

rc_hertzy · ‎08-22-2018

Thanks for your solution. I had avoided trying a FOR loop because the few steps when doing this interactively were so seductive that I was sure there was an equally simple script using BY. Your approach will definitely work and with the custom group titles will allow me to copy/paste all the final graphs into my document in one step without having to format/resize each graph by hand.

rc_hertzy · ‎08-21-2018

Goal: To script Bivariate plots of three Ys using a BY variable as the outer loop. Interactively it is easy: Fit Y by X. Select the three Y columns, Select X. Select the BY variable, in my case "Species". The resulting Bivariate plots are grouped with the Y(1), Y(2), Y(3) by X plots together for each BY value. Scripting: I only found inserting BY within each platform call. My try is: Bivariate( Y( :Name( “Y(1)” ) ), X( :X ), By( :Species) ), Bivariate( Y( :Name( “Y(2)” ) ), X( :X ), By( :Species) ), Bivariate( Y( :Name( “Y(3)” ) ), X( :X ), By( :Species) ), <<{Arrange in Rows( 1 )} But the grouping is now Y(1) vs X for all values of Species, then Y(2) for all values of Species, etc. How can I put the BY variable as the outer loop? The script created from the interactive version is not general, i.e., instead of a BY variable, it explicitly uses WHERE (:Species == "rat"), etc. The options for the BY variable change with each data set so cannot be explicitly declared using WHERE.

rc_hertzy · ‎06-20-2018

Some great changes. Is Projects available for Mac users? Previously it was only for Windows versions.

rc_hertzy · ‎06-09-2017

Axis lines in previous JMP versions for overlay plots used to be fairly visible, black and thickness of 2, I recall. Now those axis lines (and tick marks) are gray and very thin. Official response to questions was that it was changed to focus more on what was graphed instead of the background frame/structure. Graphs are for communication, so easily visible axis lines are important. Per user suggestions, one work-around is add a reference line around 0 on x-axis (to draw on top of y-axis) and simliarly on y-axis (to draw on top of x-axis). That helps a little but does not help show tick marks and does not stay with the axis if one changes the axis range. Please fix this. Many of us now use JMP for presentation graphs and the default lines are way too faint to be useful.

rc_hertzy · ‎07-03-2014

Almost correct: the Concatenate statement should be dt3 = dt2 << Concatenate ( dt1 ) ;

rc_hertzy · ‎11-14-2013

Here is a manual way. I used it to overlay a 45 degree line onto a regression plot but also works for your bubble plot. First, make your bubble lot and then separately obtain your regression line. I assume you do not want the raw data so be sure to click off Show Points in the Bivariate Fit red triangle. Then in that regression line graph, rt-clk Edit:Copy Frame Contents, then click on the target graph (bubble plot) to make it active, and rt-clk Edit:Paste Frame Contents. That places your regression line (contents of the regression frame) onto the bubble plot frame. You need to have the Arrow active, not the select (cross) tool, when right clicking. Interesting that at least as of JMP10, the two graphs do not need to be the same size. If the x-axes are not the same, the line is still accurate but might not fill the space: if the target x-axis is longer (wider) than the regression line plot, the line only goes to its original x-width. I have not found a way to script this yet.

rc_hertzy · ‎08-13-2012

Bubble plots are so different from a regression plot. Is there any way to overlay both onto the same graph? I realize one must lose the animation slider and any other differences between the two, but both are Y vs X so there should be some way to do this. Any ideas?

rc_hertzy · ‎01-07-2011

If you are considering ANOVA, then use One Way (e.g., from the JMP Starter window) on the data and select Display Options>Box Plots, and Display Options>Connect Means. It treats the x axis as categories, so if a continuous variable, the spacing is not correct. In this example with dose-response data, Dose is the X,Grouping variable. What I like is then the Oneway window is open so you can explore other analyses (e.g., equal variances). Here is a script: New Window( "Dataset: Oneway of Response by Dose", H List Box( Oneway( Y( :Response ), X( :Name( "Dose" ) ), Box Plots( 1 ), Connect Means( 1 ), Mean Diamonds( 0 ), Points Jittered( 1 ), Grand Mean( 0 ) ), ) );

rc_hertzy · ‎10-06-2010

I often encounter publications that only report means and std errors, so the option of using raw data does not exist. When I doubt their statistical conclusions I would like to run my own analyses. Can JMP use such data in ANOVA and in testing for equal variances?

Date Registered	‎06-23-2011 08:01 AM
Date Last Visited	16 hours ago
Total Messages Posted	18
Total Kudos Received	5

Online Status	Offline
Date Last Visited	16 hours ago

Re: Nonlinear modeling, part 4

Re: A simple way to automate "File Import/Append/Cleaning/Generate Report"

Re: Is it possible to create an overlay plot representing confidence intervals?

Re: Simple way to convert multiple .jmp files into .csv files (no modification) ...

Re: Arrange contour plot in groups

Re: HTML5 Auto-Publishing

Re: Concatenate Tables and close the input files

Re: Concatenate Tables and close the input files

Re: Fit curve in JMP13.0- is plot scriptable?

Re: Scripting platform with BY variable as outer loop

Nonlinear modeling, part 4

Simple way to convert multiple .jmp files into .csv files (no modification) in 2...

Re: Overlaying multiple curve fits into one plot

Re: Arrange contour plot in groups

Re: Stay organized with JMP 14

Add option to set thickness of axis lines (JMP Pro 13, Macintosh)

Re: Nonlinear modeling, part 4

Re: A simple way to automate "File Import/Append/Cleaning/Generate Report"

Re: Is it possible to create an overlay plot representing confidence intervals?

Re: Simple way to convert multiple .jmp files into .csv files (no modification) ...

Re: Arrange contour plot in groups

Re: HTML5 Auto-Publishing

Re: Concatenate Tables and close the input files

Re: Concatenate Tables and close the input files

Re: Fit curve in JMP13.0- is plot scriptable?

Re: Scripting platform with BY variable as outer loop

Scripting platform with BY variable as outer loop

Re: Stay organized with JMP 14

Add option to set thickness of axis lines (JMP Pro 13, Macintosh)

Re: How drop scripts and notes when concatenating data tables?

Re: How can I overlay bubble plot with regression?

How can I overlay bubble plot with regression?

Re: Box plot mean point

Re: analyzing average +/- st error data: common with published articles