turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- How to test for independent data using JMP?

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

Highlighted

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Aug 17, 2017 2:34 AM
(2684 views)

My customer asked me to test a few test of data for independent data. He told me to use Run Chart and check p value. However i am unable to determine how this is done.

Basically what we are trying to achieve is take data sets for a particular characteritsics from many machines,

Our intent is to check the following, We should test for normality, independency and hypothesis testing for mean and equal variances. I am able to do all the test except independency. Any help to solve this will be useful.

2 ACCEPTED SOLUTIONS

Accepted Solutions

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Sometimes when I hear about independence of runs, the concern has to do with autocorrelation. That's easy to calculate, but even easier to turn on in the Distribution Platform, (summary statistics, red triangle, Customize summary statistics, turn on autocorrelation.

JMP Systems Engineer, Pharm and BioPharm Sciences

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

To pile onto my colleague @Byron_JMP you may also want to take a look at the Specialized Modeling -> Time Series platform if indeed your data is in time series form by examining autocorrelation at lag time t.

5 REPLIES

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

There is an JMP Addin in the File Exchange that will calculate the Run's Test you are looking for.

Just download and install the Adding, and you can then run it.

Jim

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Sometimes when I hear about independence of runs, the concern has to do with autocorrelation. That's easy to calculate, but even easier to turn on in the Distribution Platform, (summary statistics, red triangle, Customize summary statistics, turn on autocorrelation.

JMP Systems Engineer, Pharm and BioPharm Sciences

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Aug 25, 2017 2:36 AM
(2564 views)
| Posted in reply to message from Peter_Bartell 08/17/2017 04:22 PM

What is the metric we are looking for in the time series or autocorrelation to determine if its independent data or dependent.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

I'm not big on using any single 'metric' to evaluate autocorrelation within time series data when one has the rich set of data visualization options within JMP. Within the Analyze -> Specialized Modeling -> Time Series platform report the Time Series Basic Diagnostics autocorrelation and partial autocorrelation plots provide lots of useful information such as whether or not autocorrelation is likely to exist...and then if it's apparent, at what lag? A great way to see how the platform works is to run the embedded analysis script in the JMP data table, 'Seriesg' (in the JMP Sample Data Directory).

Here's some additional information within the JMP online documentation:

http://www.jmp.com/support/help/13-2/Time_Series_Basic_Diagnostics_Chart.shtml#399360