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
- :
- Moving Range Chart UCL calculation in JMP8

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

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

Moving Range Chart UCL calculation in JMP8

Jun 6, 2010 8:39 PM
(1406 views)

The Moving Range UCL calculation is given in text books as UCL=D4(MRavg) with D4=3.267. When I set the JMP control chart IR at 3 Ksigma the UCL for the MR agrees with the calculation above. When I set the JMP control chart IR at 3.5 Ksigma I get a different UCL for the Moving Range chart. I understand the manual calculation for the accompanying Individuals Chart for 3.5 Ksigma (UCL=Xavg +/-3.5xMRavg/1.128) but what is the manual calculation behind the MR UCL for 3.5 Ksigma. I can't figure it out based on the different limits using 3 and 3.5 Ksigma. I need to know this for validation reasons. Thanks.

2 REPLIES

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

Re: Moving Range Chart UCL calculation in JMP8

D4 = 1 + k*d3 / d2 where d2 and d3 are unbiasing constants, dependent on sample size used to calculate the range. D4 is often tabled for k=3. For other values you can use the formula provided you have values for d2 and d3. For a moving range based on 2 samples, d2 = 1.128 (as you already know), and d3 = 0.8525. If you calculate D4 using these values and k=3.5 your calculated moving range UCL will agree with JMP.

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

Re: Moving Range Chart UCL calculation in JMP8

Thank you very much for this! I could not find this information anywhere! Much appreciated!