About bwanders

bwanders · ‎09-19-2017

Thank you. I'm rockin' and rollin' now!

bwanders · ‎09-19-2017

I have a hard time with conditional formatting. In the attached file, for example, I would like to assign pass/fail codes, 0 or 1, in column 2 depending on the value in column 1. How would I set up the formula for, If Column 1 > 0.5, then Column 2 = 1, If column 1 <0.5, then column 2 = 0?

bwanders · ‎08-05-2015

Use the "nonlinear" platform under "analyze" and "modeling". After fitting curves, you can use the "Custom Inverse Prediction" for interpolating "x" values.

bwanders · ‎08-03-2015

Thank you for your help.

bwanders · ‎07-29-2015

But after I adjust these limits, there is no constraint on values greater than zero when "None" is selected. Once "None" is selected, X1 should be constrained to 0.

bwanders · ‎07-28-2015

I have mixture data in which one of the ingredients has three categories- "Type A", "Type B", and "None". "None" wasn't part of the original DOE but was manually augmented. Is it possible to constrain the profiler settings from any number >0 when "None" is selected? Attached is an example file.

bwanders · ‎07-28-2015

Thank you, Karen.

bwanders · ‎07-27-2015

I have attached an example mixture design with one categorical factor for one of the ingredients. How would I create a formula to compute the total cost of the formula when it depends on which type of ingredient is used? If X1 is type A, the total cost is $1/kg x % X1 + $0.50/kg x % X2 + $0.75/kg x X3 If X2 is type B, the total cost is $2/kg x % X1 + $0.50/kg x % X2 + $0.75/kg x X3 If X3 is type C, the total cost is $3/kg x % X1 + $0.50/kg x % X2 + $0.75/kg x X3 Then I would be able to use the prediction formula for targeted costs.

bwanders · ‎07-23-2015

Yes, I see that I didn't have to select "Add Level". JMP automatically split the constant value into two boxes of 0.2 levels. And I did remember to change the auto sum to recognize the new value. Thank you for your help.

bwanders · ‎07-23-2015

I don't have JMP12.1 yet, so this is what I'm seeing in JMP11.2: After changing the design role to mixture and setting up the mixtures for the augment design, X4 is still a constant. But, if I select to the left of X4, I am able to "Add Level", in which it automatically selects 1.2. Now I can manually change this value and continue on with adding all 2-way interactions.

bwanders · ‎07-22-2015

I have 4 mixtures. One mixture is a constant at 0.20. I used the mixture sum option to set ingredients 1-3 to equal 0.8. Could you augment this design by changing the constant mixture to a variable and eliminating the mixture sum advanced option? I have attached this example DOE.

bwanders · ‎05-29-2015

Yes. It all makes sense to me now. Thank you Michael. I now have a more practical analysis.

bwanders · ‎05-28-2015

Originally, I started with a probability column (y/100). Look at the JMP alert I get when I try to run the model after I have chosen the logit transformation.

bwanders · ‎05-28-2015

Michael, Could you send the sample file back with the corrected transformation? I'm having trouble with it.

bwanders · ‎05-27-2015

I have attached an example mixture design with response data in percentages. When you use standard least squares, it is possible to have predictions outside of 0 and 100%. Isn't there way to constrain responses to 0 and 1 or 0 and 100?. The generalized linear model can't take mixture effects.

bwanders · ‎03-17-2015

Example- You are generating a mixture design for a new food recipe, and your response will be the proportion of customers who would order a particular food dish again.

bwanders · ‎01-29-2015

"use of a Taguchi L9 in a mixture problem is really a misapplication of DOE or at the very least a poor fit of DOE principles in a mixture context"- joe.n.hockman I completely agree with your statement above, Joe.

bwanders · ‎01-29-2015

Thank you, Dan. I now have enough information to back up what I am trying to teach. It would be so much easier to start out with the right design in the first place.

bwanders · ‎01-27-2015

Dan, I definitely like your explanations in all three paragraphs. In most cases where I help colleagues analyze their data, the designs were not setup in JMP as a Taguchi Array design. So when I receive the JMP file, all mixtures are designated as continuous variables in which the sum totals 100%. I have compared the regression analysis when all mixture factors are continous to when they are properly set to the mixture role. This way I can show the difference in predictions using the profiler. Now what would you say to the people that are confused to why the prediction profiler is not showing curvature?

bwanders · ‎01-26-2015

Correction- colleagues

bwanders · ‎01-26-2015

Some of my collaegues are using Taguchi Arrays for their mixture DOE's. In your opinion, what are the biggest mistakes you can make by using an L9 design for 3 ingredients at 3 levels instead of using a proper mixture DOE?

bwanders · ‎12-24-2014

Julian, Thank you for the Christmas present. I hope you have an enjoyable holiday.

bwanders · ‎12-19-2014

Is there any chance I could get the script for altering the Dice Rolls simulation under the JMP Teaching Resources to have 2 or 3 dice rolls and plot the mean instead of just one roll?

bwanders · ‎07-21-2014

I have attached an 8 mixture custom DoE in which 3 ingredients are constant. When I run the model, the prediction profiler predicts the maximum Y1 & Y2 within reason of the actual maximum. But when I minimize the desirabilities, I get negative values. How is it getting these numbers and how would I prevent any predictions falling well below the actual low response values?

bwanders · ‎03-06-2014

Thank you. This nonlinear custom model definitley brought me to a new place in JMP.

bwanders · ‎03-05-2014

Click on JMP in the Excel toolbar, then click on "preferences". You will see where to check the box for "Use first rows as column names"

bwanders · ‎03-05-2014

Click on JMP in the Excel toolbar, then click on "preferences". You will see where to check the box for "Use first rows as column names"

bwanders · ‎03-04-2014

Look at my attached file. I have fitted an Exponential 3P curve to data that is based on %. I have added a range check as well as a response limit for my "% knockdown" column. I would like to have my fit curve parameter estimates not exceed 100%. Is there a way to do this?

bwanders · ‎02-28-2014

In quotes- that's what I was missing. Thank you Karen.

bwanders · ‎02-27-2014

In the "Nonlinear" fit curve graph, how do I change the line color?

Date Registered	‎09-20-2013 12:32 PM
Date Last Visited	‎11-05-2019 01:05 PM
Total Messages Posted	30
Total Kudos Received	4

Online Status	Offline
Date Last Visited	‎11-05-2019 01:05 PM