BookmarkSubscribe
Choose Language Hide Translation Bar
vestlink
Community Trekker

Define classes in Fit Model

Hi.

 

in SAS you can defines class in a script.

 

How do I do the same in JMP scripting?

 

Fit Model(
Y( :korr_avvendingsvekt ),
Effects(
:parity_recode 2,
:TVILLING,
:Beitedager,
:beiteklasse,
:rasetype,
:beiteklasse * :rasetype,
:Sex,
:tetthet
),
Random Effects( :mor_ID_cat, :kalvings_year_cat ),
Keep dialog open( 1 ),
Personality( "Mixed Model" )
)

 

 

0 Kudos
1 ACCEPTED SOLUTION

Accepted Solutions
txnelson
Super User

Re: Define classes in Fit Model

The simplest way in JMP to find out what the JSL to use for a given platform, is to interactively setup the platform the way you want it, and then to have JMP provide you with the script, by going to the red triangle on the output window, and selecting "Save Script".  JMP uses the Modeling Type of the columns specified in the Effects specification to determine which of the columns are Continuous and which are Class(Ordinal or Nominal) effects.  Documentation on this can be found at:

     Help==>Books==>Fitting Linear Models

fitmixed.PNG

Fit Model(
	Y( :Y ),
	Effects(
		:Treatment,
		:Month,
		:Treatment * :Month,
		:Name( "AM/PM" ),
		:Treatment * :Name( "AM/PM" ),
		:Month * :Name( "AM/PM" ),
		:Treatment * :Month * :Name( "AM/PM" )
	),
	Center Polynomials( 0 ),
	Personality( "Mixed Model" ),
	Subject( :Patient ),
	Repeated Effects( :Time ),
	Repeated Structure( "Unstructured" ),
	Run( Random Effects Covariance Parameter Estimates( 0 ) )
);

 

Jim
0 Kudos
1 REPLY 1
txnelson
Super User

Re: Define classes in Fit Model

The simplest way in JMP to find out what the JSL to use for a given platform, is to interactively setup the platform the way you want it, and then to have JMP provide you with the script, by going to the red triangle on the output window, and selecting "Save Script".  JMP uses the Modeling Type of the columns specified in the Effects specification to determine which of the columns are Continuous and which are Class(Ordinal or Nominal) effects.  Documentation on this can be found at:

     Help==>Books==>Fitting Linear Models

fitmixed.PNG

Fit Model(
	Y( :Y ),
	Effects(
		:Treatment,
		:Month,
		:Treatment * :Month,
		:Name( "AM/PM" ),
		:Treatment * :Name( "AM/PM" ),
		:Month * :Name( "AM/PM" ),
		:Treatment * :Month * :Name( "AM/PM" )
	),
	Center Polynomials( 0 ),
	Personality( "Mixed Model" ),
	Subject( :Patient ),
	Repeated Effects( :Time ),
	Repeated Structure( "Unstructured" ),
	Run( Random Effects Covariance Parameter Estimates( 0 ) )
);

 

Jim
0 Kudos