The Elsmar Cove Wiki More Free Files The Elsmar Cove Forums Discussion Thread Index Post Attachments Listing Failure Modes Services and Solutions to Problems Elsmar cove Forums Main Page Elsmar Cove Home Page
Google
  Web Elsmar.com
*Please be aware that SOME RECENT forum threads may not yet be indexed by Google.

View Full Version : Control Chart for Dependent obs - One factor for each category weekly


mmjj3259
4th June 2009, 04:13 PM
Thanks a lot!!:lol:

Bev D
5th June 2009, 02:00 PM
the default generic answer is an X, MR chart given single values, assuming continuous data. However, depending on the type of dependence it may not be the correct choice at all. Without more information about the processes and 'factors' it is impossible to say.

The one firm comment is that the control limits should not constantly readjusted. They should be set based on the stable period and left alone until a permanent chagne for the good has occured.

mmjj3259
8th June 2009, 12:05 PM
Do all types of control charts assume that all the obs are independent?

Bev D
8th June 2009, 12:08 PM
no. but all of the standard, commonly known & taught charts do.

mmjj3259
8th June 2009, 01:10 PM
no. but all of the standard, commonly known & taught charts do.


Any ideas of how to deal non-independent obs?

Bev D
8th June 2009, 01:26 PM
sure. lots of them but without more details about what you plotting (what is the process? what is the output you are plotting, what are the factors you mentioned?) and maybe even seeing your data, I can make no specific recommendations.

mmjj3259
9th June 2009, 01:53 AM
sure. lots of them but without more details about what you plotting (what is the process? what is the output you are plotting, what are the factors you mentioned?) and maybe even seeing your data, I can make no specific recommendations.

This is 104 weeks data.
0.62630.62450.63160.63280.63030.62310.61870.62010.62580.62460.62170.62070.61720.61880.62030.62150.61890.62310.62420.63470.63340.62630.62350.6180.61830.63490.62740.62860.61710.61720.61080.61440.6080.60540.6010.59580.61810.60990.59590.60460.60510.60270.60160.60940.60380.60950.60950.62560.62940.62040.61640.61470.61350.60840.61010.61280.610.60850.59870.60580.60830.61070.61160.62630.62580.62980.63310.64140.65380.64670.65130.65680.65830.65620.66250.66620.66490.67170.67020.66920.66940.65790.6580.66630.65970.64690.6340.62560.63210.6230.62040.61640.61920.61180.61070.62250.61160.62180.63650.63650.64050.63040.63680.6397

For example the first obs is calculated by number 1 and number2, the second obs is calculated by number2 and number3, the third obs is calculated by number3 and number4. Do we say they are non-independent?

Bev D
9th June 2009, 07:16 AM
I'm sorry you really need to be more specific.
what is this data? what is the 'calculation' that creates each observation? are the numbers you pasted the 'calculated number' or the 'observations'? We cannot help you if you do nto answer these basic questions.