Effect estimation in GLM

[mawais31]

Hi,

I have Generalized Linear model with log(sales) = mileage + ageofcar +  temperature + precepitation + summer + spline(weekday) + spline(day) + spline (time) as my model

I got a deviance of 656 and I got all variables as significant

I mileage, ageofcar, summer as categorical variables

I want to find the effect of predictors on my response variable , is there any way to find it

Thanks in Advance

[Maguin, Eugene]

Since you haven't posted your syntax, I'm not sure but I think you should look at the documentation, i.e., the syntax reference, and note that the subcommand, Print parameter, will cause the coefficients to be printed, which, I think, is what you want to see.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of mawais31
Sent: Wednesday, October 10, 2012 7:41 AM
To: [hidden email]
Subject: Effect estimation in GLM

Hi,

I have Generalized Linear model with log(sales) = mileage + ageofcar + temperature + precepitation + summer + spline(weekday) + spline(day) + spline (time) as my model

I got a deviance of 656 and I got all variables as significant

I mileage, ageofcar, summer as categorical variables

I want to find the effect of predictors on my response variable , is there any way to find it

Thanks in Advance



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Effect-estimation-in-GLM-tp5715570.html
Sent from the SPSSX Discussion mailing list archive at 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

=====================
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

[mawais31]

Thanks Gene Maguin for quick reply!


Probebly the question seems quite basic, but right now I am stuck..

Here is code/syntax!


<code>
ods html;
ods graphics on;

proc gam data=WORK.ACCIDENT2010;
class mileage age_of_car companycar summer;
freq accident;

model log_SALES = spline(time, df=20) spline(month, df=5) spline(weekday) spline(day, df=10) param(mileage age_of_car ttn companycar prr summer);
output out=WORK.ACCIDENTOUT pred resid;
run;
proc gplot data=WORK.ACCIDENTOUT;
plot log_SALES * time = 1
         P_log_SALES * time = 2
         R_log_SALES * time = 3/overlay ;

symbol1 v=dot c=blue i=none;
symbol2 v=dot c=red i= none;

title 'log_Sales VS time';
run;

ods graphics off;
ods html close;
</code>

I got the parameter values but how I will interpret it if parameter for mileage is .12, so can I say that if there is an increase of 100 cars in mileage there there is probability of 12% increase in sales...
and my temperature value is -.0029...

Where ttn and prr is temperature and precepetation respectively

Awais khan

[Bruce Weaver]

You've got the wrong forum.  That is SAS code.  This is an SPSS mailing list.  Try comp.soft-sys.sas or SAS-L.  

mawais31 wrote

Thanks Gene Maguin for quick reply!


Probebly the question seems quite basic, but right now I am stuck..

Here is code/syntax!


<code>
ods html;
ods graphics on;

proc gam data=WORK.ACCIDENT2010;
class mileage age_of_car companycar summer;
freq accident;

model log_SALES = spline(time, df=20) spline(month, df=5) spline(weekday) spline(day, df=10) param(mileage age_of_car ttn companycar prr summer);
output out=WORK.ACCIDENTOUT pred resid;
run;
proc gplot data=WORK.ACCIDENTOUT;
plot log_SALES * time = 1
         P_log_SALES * time = 2
         R_log_SALES * time = 3/overlay ;

symbol1 v=dot c=blue i=none;
symbol2 v=dot c=red i= none;

title 'log_Sales VS time';
run;

ods graphics off;
ods html close;
</code>

I got the parameter values but how I will interpret it if parameter for mileage is .12, so can I say that if there is an increase of 100 cars in mileage there there is probability of 12% increase in sales...
and my temperature value is -.0029...

Where ttn and prr is temperature and precepetation respectively

Awais khan

[Genevieve Odoom]

Thanks for your email. I will be out of the office on Thursday October 11th, returning on Friday, October 12th with limited access to email. I will respond to all emails when I return.

Genevieve Odoom
Policy and Program Analyst
OANHSS
Suite 700 - 7050 Weston Rd. Woodbridge,
ON L4L 8G7
Tel: [phone] x 241 Fax: [phone]
[hidden email]
 www.oanhss.org<https://mail.oanhss.org/ecp/Organize/www.oanhss.org>

=====================
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