logistics model.
in [SAS]
|
From: Murphy Choy on 21 Oct 2009 20:41 Hi, You can score the dataset by including it in a score statement in the original proc logistic. ------Original Message------ From: Jeff Sender: SAS(r) Discussion To: SAS-L(a)LISTSERV.UGA.EDU ReplyTo: Jeff Subject: logistics model. Sent: Oct 22, 2009 8:30 AM I used dataset A to develop logistic model for predictive purpose (contains about 30 predictors) I want to predict dataset B, what is the best way to do that Thank, Jeff Sent from my BlackBerry Wireless Handheld -- Regards, Murphy Choy Certified Advanced Programmer for SAS V9 Certified Basic Programmer for SAS V9
From: Wensui Liu on 21 Oct 2009 20:54 what murphy said is very true. however, i had bad experience with the sas score file. the best way per my experience is to write the formula out in an explicit functional form. On Wed, Oct 21, 2009 at 8:41 PM, Murphy Choy <goladin(a)gmail.com> wrote: > Hi, > > You can score the dataset by including it in a score statement in the original proc logistic. > > ------Original Message------ > From: Jeff > Sender: SAS(r) Discussion > To: SAS-L(a)LISTSERV.UGA.EDU > ReplyTo: Jeff > Subject: logistics model. > Sent: Oct 22, 2009 8:30 AM > > I used dataset A to develop logistic model for predictive purpose (contains > about 30 predictors) > I want to predict dataset B, what is the best way to do that > Thank, > > Jeff > > > Sent from my BlackBerry Wireless Handheld > > -- > Regards, > Murphy Choy > > Certified Advanced Programmer for SAS V9 > Certified Basic Programmer for SAS V9 > -- ============================== WenSui Liu Blog : statcompute.spaces.live.com Tough Times Never Last. But Tough People Do. - Robert Schuller ==============================
From: Murphy Choy on 21 Oct 2009 20:59 Hi Wensui, Well that is typically what is required in banking sectors where the people are concerned about the scorecard discrepancies where small differences of 1 or 2 points is cause for alert. But I totally agreed with you on the fact that people can be uncomfortable with using an automated scoring system. Occasionally I do use hand counting as a form of verification. ------Original Message------ From: Wensui Liu To: goladin(a)gmail.com Cc: SAS-L(a)listserv.uga.edu Subject: Re: logistics model. Sent: Oct 22, 2009 8:54 AM what murphy said is very true. however, i had bad experience with the sas score file. the best way per my experience is to write the formula out in an explicit functional form. On Wed, Oct 21, 2009 at 8:41 PM, Murphy Choy <goladin(a)gmail.com> wrote: > Hi, > > You can score the dataset by including it in a score statement in the original proc logistic. > > ------Original Message------ > From: Jeff > Sender: SAS(r) Discussion > To: SAS-L(a)LISTSERV.UGA.EDU > ReplyTo: Jeff > Subject: logistics model. > Sent: Oct 22, 2009 8:30 AM > > I used dataset A to develop logistic model for predictive purpose (contains > about 30 predictors) > I want to predict dataset B, what is the best way to do that > Thank, > > Jeff > > > Sent from my BlackBerry Wireless Handheld > > -- > Regards, > Murphy Choy > > Certified Advanced Programmer for SAS V9 > Certified Basic Programmer for SAS V9 > -- ============================== WenSui Liu Blog : statcompute.spaces.live.com Tough Times Never Last. But Tough People Do. - Robert Schuller ============================== Sent from my BlackBerry Wireless Handheld -- Regards, Murphy Choy Certified Advanced Programmer for SAS V9 Certified Basic Programmer for SAS V9
|
Pages: 1 Prev: Netezza - hide password Next: Quality of logistic regression model |