Date:         Sat, 16 Dec 2006 04:39:01 -0500
Reply-To:     Marina Kekrou <mkekrou@YAHOO.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Marina Kekrou <mkekrou@YAHOO.CO.UK>
Subject:      Re: Jacknife macro

Ian,

Thanks for this. As I noted in a previous email I am want to get out of
sample prediction for observation i without including this observation.
Is this equivalent to what you are suggesting? What is the difference then
between that and jacknife?

thanks
marina

On Fri, 15 Dec 2006 16:59:19 -0000, Ian Wakeling
<ian.wakeling@HANANI.QISTATS.CO.UK> wrote:

>Marina,
>
>Check out the documentation for the output statement
>in proc reg.
>
>If all you want are the predictions when cases are left
>out of the regression then there is no need to run your
>macro at all.
>
>Just do something like
>
>PROC REG data=test;
>  MODEL y = x z;
>  output out=regout p=ypred press=r_omited;
>RUN ;
>quit;
>
>p=ypred gives you the normal predictions in the
>output data set called regout.  press=r_omited gives you
>the residual without the current observation.  To work
>back to the prediction without the current observation
>run a simple datastep after the PROC REG
>
>data regout;
>  set regout;
>  jkpred=y-r_omited;
>run;
>
>Ian.
>
>
>----- Original Message -----
>From: "Marina Kekrou" <mkekrou@YAHOO.CO.UK>
>To: <SAS-L@LISTSERV.UGA.EDU>
>Sent: Friday, December 15, 2006 3:58 PM
>Subject: Jacknife macro
>
>
>> Any ideas on how i can modify this code to get predicted values when
>> running prog reg?
>>
>> thanks
>>
>> %MACRO JackReg ;
>> %DO I = 1 %TO &ncases ;
>> DATA temp&I ;
>>   SET one;
>>   IF  _N_   NE   &I ;
>> RUN;
>>
>> PROC REG outest =loopIest;
>>      Omits&I:  MODEL y = x z;
>>   *Specify your model in the line above;
>> RUN ;
>>
>> PROC APPEND BASE = RegEsts  NEW = loopIest ;
>> RUN ;
>>
>> %END ;
>> %MEND JackReg;
>> *Macro portion of program ends here;
>>
>> %JackReg;
>>