I would like to know if JMP is capable of computing an Extended Autocorrelation Function. I understand the EACF is useful for identifying mixed ARIMA models.
Hello -- The documentation for the Time Series platform is available here: Time Series Analysis JMP supports ARIMA and Seasonal ARIMA models, as well as Autocorrelation and Partial Autocorrelation Time Series Commands.
ESACF is supported in SAS/ETS (SAS/ETS(R) 13.2 User's Guide).
You might also consider using the JMP interface to R to calculate the ESACF (TSA package).
Here is a quick GUI that uses some JMP sample data to compute the sample extended acf.
Returns the following to the log:
.jsl script is attached. Note ar.max and ma.max are not parameterized right now but easily could be added to the GUI as user input.
Thank you Daniel. Unfortunately, I don’t know how to develop a GUI. Can
you point me in a direction where I can do some research on the topic? I
have looked on the JMP user Community. I couldn’t find anything there.
Sent: Monday, May 04, 2015 3:49 PM
To: Desta Ledom
Subject: Re: - I would like to know if JMP is capable of
computing an Extended Autocorrelation Function. I understand the EACF is
useful for identifying mixed ARIMA models.
JMP User Community
would like to know if JMP is capable of computing an Extended
Autocorrelation Function. I understand the EACF is useful for identifying
mixed ARIMA models.
reply from danielvalente
Discussions - View the full discussion
My pleasure -- You can start with my script as a template, but one resource I would recommend checking out is "Jump in JMP Scripting" by Murphrey and Lucas. Link to the Book: Jump into JMP® Scripting
Also, the scripting index in JMP is quite rich. You can find that in Help > Scripting Index in the product.
Also in our documentation, there is a nice index that will help give you some basics on JSL: Help with Learning JSL
Just as an aside, you don't need to make a GUI to make this script work, you can run it all from a script editor if you just need to process a data set with the R function. The GUI just helps to make it more generalizable and "JMP-like".
Hope this helps!