Process Screening Drift Summaries

[#1]

Hi All,

Are there any more documentations available for the outputs generated from the Process Screening Report. In particular, I would like to understand how JMP calculates the Drift Summary values. I am aware that they are using Holt's Double EWMA Smoother on the parameters but there is no explanation on the Drift Summary values. Trying to understand the math behind before attempting to use it to screen through a large number of variables to determine if any of them are drifting "significantly".

Thanks! 

[#2]

You should contact JMP Support for the answer on this.  

     support@jmp.com

The have access to the developers

[#3]

Okay thanks. Will give that a try.

[#4 · Accepted solution]

Chris,

The calculation of the drift values in Process Screening is described in more detail in this paper by John Sall: Scaling-up process characterization (starting at the bottom of the page numbered 73). A further reference (also given in the aforementioned paper) is this book by Hyndman, et al:

Hyndman, Koehler, Ord, and Snyder. 2008. Forecasting with exponential smoothing, the state space approach, Berlin: Springer Verlag.

[#5]

Hi sseligman,

Thanks for the document! This should be what I was looking for.