Subscribe Bookmark RSS Feed

SAS Proc GLM in JMP

gutloja

Community Trekker

Joined:

Jul 31, 2015

Hello, All

I need to re-run a SAS script in JMP, but I do not know how to add a column as co-variable. 

This is the SAS script:

DATA EXPERIMENT;

INPUT trat rep Initial a b c Response;

Cards;

PROC GLM DATA=EXPERIMENT;

    class N P K;

    model Response = a |b|c|Initial/solution;

    LSMEANS a |b|c|/stderr PDIFF ADJUST=TUKEY;

    OUTPUT OUT= COVAR R=RES P=PREDICTED;

PROC UNIVARIATE data = COVAR normal;

    VAR RES;

QQPLOT RES/normal (mu=est sigma=est);

PROC GPLOT EXPERIMENT = COVAR;

Plot RES*PRED=PRED/VREF=0;   

    RUN;

    QUIT;


And this is the script I am using in JMP, but the results look different.

Fit Model(

    Y( :Response ),

    Effects( :a, :b, :a * :b, :c, :a * :c, :b * :c, :a * :b * :c ),

    Personality( "Generalized Linear Model" ),

    GLM Distribution( "Normal" ),

    Link Function( "Identity" ),

    Overdispersion Tests and Intervals( 0 ),

    Name( "Firth Bias-adjusted Estimates" )(0),

    Run

)

Any help will be appreciated.


Thank you,


JoseL

1 REPLY
jiancao

Staff

Joined:

Jul 7, 2014

What part of results are different? Coefficient estimates? If you can provide specifics and data used, I will take a look at it.