GEE analysis linear combination

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

GEE analysis linear combination

Hersenvocht
This post was updated on .
Hi all,

I need to test a linear combination of coefficients (or planned contrast) in
the following model:
beta1 x var1 + beta2 x var2 + beta3 x var1 x var2
I need to test beta1+beta3 < -beta1 or alternatively 2 x beta1 + beta3 <0
I cannot figure out how to test combinations of coefficients in the GEE
function (or the equivalent GLM)
Is there anyone that can help me out?

Thanks in advance!

Excuse me for not including:
var1 is a discrete variable from 0 to 9 and var2 is a dummy variable which can be 0 or 1


--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Reply | Threaded
Open this post in threaded view
|

Re: GEE analysis linear combination

Bruce Weaver
Administrator
You said:  "var1 is a discrete variable from 0 to 9"

In that case, what you called beta1 is really a set of 9 coefficients, is it
not?



Hersenvocht wrote

> Hi all,
>
> I need to test a linear combination of coefficients (or planned contrast)
> in
> the following model:
> beta1 x var1 + beta2 x var2 + beta3 x var1 x var2
> I need to test beta1+beta3 < -beta1 or alternatively 2 x beta1 + beta3 <0
> I cannot figure out how to test combinations of coefficients in the GEE
> function (or the equivalent GLM)
> Is there anyone that can help me out?
>
> Thanks in advance!
>
> Excuse me for not including:
> var1 is a discrete variable from 0 to 9 and var2 is a dummy variable which
> can be 0 or 1
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: GEE analysis linear combination

Hersenvocht
No, the variable can take on the value 0,1,2,3,4,5,6,7,8,9. If I include the
variable in my GEE regression as covariate I just get one coefficient, as I
like.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Reply | Threaded
Open this post in threaded view
|

Re: GEE analysis linear combination

Bruce Weaver
Administrator
Thanks for clarifying.  So if I follow, you are using GENLIN to estimate your
model, and the problem is that GENLIN does not have LMATRIX or TEST options
to make the comparisons of interest.  Is that right?  


Hersenvocht wrote

> No, the variable can take on the value 0,1,2,3,4,5,6,7,8,9. If I include
> the
> variable in my GEE regression as covariate I just get one coefficient, as
> I
> like.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.