# [R] glm fit

Richardson, Patrick Patrick.Richardson at vai.org
Fri May 8 13:58:11 CEST 2009

```Interesting. Now the fifth homework question you have posted here. As was stated before (by several members) this is not a homework forum.

Patrick

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of mathallan
Sent: Friday, May 08, 2009 5:50 AM
To: r-help at r-project.org
Subject: [R] glm fit

Hi, I try to ask here, because I hope someone will help me understand this
problem-

I have fittet a glm in R with the results

> glm1 <-
> summary(glm1)

Call:
glm(formula = log(claims) ~ log(sum) * as.factor(grp), family =

Deviance Residuals:
Min       1Q   Median       3Q      Max
-6.6836  -1.3626  -0.2576   1.2038   8.2480

Coefficients:
Estimate Std. Error      t value   Pr(>|t|)
(Intercept)                     3.525657   0.436102   8.084 8.18e-16 ***
log(sum)                        0.334288   0.025668  13.024  < 2e-16 ***
as.factor(grp)2               0.434262   0.976240   0.445   0.6565
as.factor(grp)3               3.666490   1.436471   2.552   0.0107 *
as.factor(grp)4               0.040782   1.024730   0.040   0.9683
log(sum):as.factor(grp)2  0.007719   0.061914   0.125   0.9008
log(sum):as.factor(grp)3 -0.209986   0.091578  -2.293   0.0219 *
log(sum):as.factor(grp)4  0.059342   0.067320   0.881   0.3781
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for gaussian family taken to be 3.693731)

Null deviance: 15839  on 4035  degrees of freedom
Residual deviance: 14878  on 4028  degrees of freedom
AIC: 16737

Number of Fisher Scoring iterations: 2

But i'm not sure what I get out of the summary. What does it tell me?

How should the formula for Y look like, out from the summary? Is it
something like

Y=0.334288*X_sum+0.434262*X_2+  ??

And when I get a expression for Y, what does it tell me? Is it an expected
expression?

--
View this message in context: http://www.nabble.com/glm-fit-tp23443121p23443121.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
and provide commented, minimal, self-contained, reproducible code.

This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential information.  Any unauthorized review, use, disclosure or distribution is prohibited.  If you are not the intended recipient(s) please contact the sender by reply email and destroy all copies of the original message.  Thank you.
```