Share your ideas for the JMP Scripting Unsession at Discovery Summit by September 17th. We hope to see you there!
Choose Language Hide Translation Bar
Highlighted
galactus3000
Level IV

creating a pdf (probability density function) of likely means differences as output of bivariate analyses

as an output from a means comparison during a bivariate analysis, i would like to generate a pdf plot of estimated means differences

i could create this manually by performing a means comparison or t-test with a given confidence level alpha (say 0.05) capture the corresponding confidence interval for the means difference to a table and then repeat the exercise for a range of alphas say 0.01, 0.025, 0.05, 0.075, 0.1, 0,125 etc in order to sweep out a pdf of means differences

does this feature exist or would a jsl script (possibly add-in) need to be written?

ultimate output would be a pdf plot perhaps

 

2 REPLIES 2
Highlighted

Re: creating a pdf (probability density function) of likely means differences as output of bivariate analyses

You can use the bootstrap feature or the simulation feature of JMP Pro with the Bivariate platform. Use the bootstrap feature if you want to resample the original observations. Use the simulation feature if you have a formula with one or more stochastic contributions.

 

These features are documented well in the Help system.

Learn it once, use it forever!
Highlighted

Re: creating a pdf (probability density function) of likely means differences as output of bivariate analyses

Here is a simple illustration of how to use the bootstrap feature. I opened the sample Big Class data table. I launched the Oneway platform with Y = weight and X = sex. I performed the ANOVA.

 

Screen Shot 2019-12-15 at 7.12.40 AM.png

 

I right-clicked on the table with the difference between the means of the two sexes and selected Bootstrap. I requested the default number of samples, 2500. I used the saved Distribution table script in the data table with the bootstrap results.

 

Screen Shot 2019-12-15 at 7.12.56 AM.png

Learn it once, use it forever!
Article Labels

    There are no labels assigned to this post.