cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar

Calculation of Tolerance Intervals for Non-Normally Distributed Data

Status: Acknowledged Submitted by on ‎01-28-2022 08:43 AM
6 Comments (6 New)

Like in other statistical software it would be nice to be able to calculate tolerance intervals for the following non-normal distributions:

  1. Lognormal
  2. Gamma
  3. Exponential
  4. Smallest extreme value
  5. Largest extreme value
  6. Logistic
  7. Loglogistic
  8. Weibull

I know there are some workarounds and normalizing via transformation ist always a possibility, but having just to press one button would be perfect.

 

Best regards,

Christian

6 Comments
Sarah-Sylvestre
Staff
Staff
‎12-08-2022 02:33 PM
‎12-08-2022 02:33 PM
Status changed to: Acknowledged

Hi @christianguhr, thank you for your suggestion! We have captured your request and will take it under consideration.

BlockMeerkat111
Level I
‎06-28-2023 10:03 AM
‎06-28-2023 10:03 AM

Piling onto this request.  New JMP user here, and I'm surprised that this feature is missing.

PatrickGiuliano
Staff
Staff
‎06-29-2023 07:31 PM
‎06-29-2023 07:31 PM

Thanks for chiming in here @BlockMeerkat111. We do have a way of generating these but not in the "traditional sense," i.e. using the Tolerance Intervals red triangle menu option under the Distribution platform (similar how you might report/ generate these in a competitors' statistical package).  I am planning to write a JMPer Cable Article that walks through that process and will try and circulate it here once I have it done.

BlockMeerkat111
Level I
‎06-30-2023 09:57 AM
‎06-30-2023 09:57 AM

Thanks, will stay tuned.  We're med dev and will have to validate the method in order to use it, so hopefully a test data set and expected results will be available with this.

ChrisGaltress
Level I
‎07-26-2023 04:25 PM
‎07-26-2023 04:25 PM

Hi Patrick,

 

Any update on JMPer Article?  Agree with others that this would be a welcome addition to JMP.  I'm currently working on a project that requires calculating tolerance intervals from numerous (100s) of simulated non-normal sample sets.  I'm working mostly w/ Weibull, Lognormal, and SHASH distributions and would like to be able to calc & export the lower bounds at 90% coverage & 95% confidence for these sample sets into table format to estimate the predicted failure rate of being below the lower bound.

PatrickGiuliano
Staff
Staff
‎08-04-2023 08:42 PM
‎08-04-2023 08:42 PM

Hello @ChrisGaltress, No update unfortunately, I haven't been able to get to this.  Let's try a different approach.  Can you please send us an email to support@jmp.com and address it to me directly? "Hi Patrick..." Please reference this community post and give the subject line a meaningful header "Calculation of Tolerance Intervals for Non-Normally Distributed Data."  I can work with you directly to demonstrate the 'alternate' approach that I am thinking for your case, and to identify its limitations (off hand, we don't support SHASH in Life Distribution which is where I think you would need to go here).