regression as effect in mixed lmodel

[Ameurgen]

hello  sas community;

 i am runing a mixed model and i need to add a variable"date of birth" as a regression effect because we dont need a single effect from date of birth but we need a trend effect for for genetic and phenotypic purpose. 

So,  if one of you have a same issue can help me.

Thank you

[PaigeMiller]

If I am understanding you properly, if you want a "trend" then you need to make date_of_birth to be a numeric variable and it should be added into the model. If it currently is not a numeric variable, you would have to create a new numeric variable via the INPUT function — but the exact details of how to do this depend on you providing an example or two of the values of date_of_birth.

[Ameurgen]

Hello Paige Miller , 

Thank for your reply, 

can you please , give me more detail how you understanding because , i dont get it , you just propose to change the format of variable to be incorporete in model as a regression effect (on effect).

find below  an example of my data

Thank you

ANM_ID	Dam_Birth_date	Dam_Date_Calving	Dam_Days_in_Milk	Dam_Lact_No	Cow_Lact_NO
9029067	16/06/2009	17/07/2011	48	1	5
9199675	25/05/2009	04/07/2011	188	1	5
9029069	16/07/2008	17/07/2011	49	2	5
9124686	13/09/2008	07/09/2011	72	2	5
9487877	14/06/2010	21/07/2012	69	1	5
9410042	19/05/2010	12/06/2012	45	1	5
9732404	08/01/2010	05/03/2013	91	2	5

[PaigeMiller]

I did not propose a change of format. I said you might need to change the variable type from character to numeric, if it is character right now. And I can't tell from your data if it is character or numeric. Can you please tell me what PROC CONTENTS says about the type of this variable?

[Ameurgen]

oh, ok,  it is a numeric variable 

Thank you

DamRegistrationId	Char	18	$18.
Dam_Birth_date	Num	8	YYMMDD10.
Dam_CLV_Mo	Num	8
Dam_Date_Calving	Num	8	YYMMDD10.
Dam_Days_in_Milk	Num	8

*

[PaigeMiller]

So you just add the date variable into your model.