Hi:
On Thu, Nov 25, 2010 at 3:53 AM, Diogo B. Provete wrote:
> I got a question about using a GLZ with categorical x categorical data.
> Below there is a data set I want to know the influence of treatments (CONT,
> and LPS2H LPS24H) on the categories of pigmentation of the right testis of
> an am
I got a question about using a GLZ with categorical x categorical data.
Below there is a data set I want to know the influence of treatments (CONT,
and LPS2H LPS24H) on the categories of pigmentation of the right testis of
an amphibian. From these data, I used the function glm with binomial family
annie Zhang wrote:
>
> Hi, Milton,
>
> Thank you for the reply. I tried, but it seems the problem is the column
> name of the test data is not the same as the column name of the training
> data. I didn't give the column name, the system seemed do. How to chang
> here?
>
> Annie
>
> On Fri, Au
I think I need to restate the problem. If the test data is only a vector,
then I am predicting one test sample. But the output from the predict result
has the same length as the training set. And there is a warning message
about this.
Annie
On Sat, Aug 8, 2009 at 2:52 PM, annie Zhang wrote:
> Hi
Hi, Milton,
Thank you for the reply. I tried, but it seems the problem is the column
name of the test data is not the same as the column name of the training
data. I didn't give the column name, the system seemed do. How to chang
here?
Annie
On Fri, Aug 7, 2009 at 7:52 PM, milton ruser wrote:
Hi Annie,
create a new data.frame with input variables having all predictors variables
on it.
after give a look at ?predict
best wishes
milton
On Fri, Aug 7, 2009 at 8:19 PM, annie Zhang wrote:
> Hi, R users,
>
> I am trying to use glm to do logistic regression. I know generally when I
> have
Hi, R users,
I am trying to use glm to do logistic regression. I know generally when I
have two covariates, say x1 and x2, then I do
fit <- glm(y~x1+x2,famliy='binomial')
But now my covariates form a n*p matrix, say x, so actually each column is a
covariate. So I think I should do
fit <- glm(y~x,f
Although both are known now, there is a time element involved in
which one, max.loss was fixed at the time of underwriting and loss was
unknown at that time. This *is* an insurance question is it not?
Wouldn't the question be: Can one use the group variable to estimate
the proportion of ma
Actually both max.loss and loss are known values (in dollars). I'm very much
doubt, what to choose.
glm(max.loss~loss,family=gaussian(link="identity")
or
glm(formula = sum ~ claims * as.factor(grp), family = gaussian(link =
"identity"))
or
glm(loss~max.loss,family=gaussian(link="identity")
w
I think you are off-track because max.loss does not sound like a
proper Y variable. Because max.loss is an amount that is known, in the
insurance applications I have seen it would have been modeled within
an offset term. Many of the examples have used number of ships or
buildings or the per
Hi
I got a dataset
loss max.loss grp
1 10 50 2
2 15 33 1
3 18 49 2
4 33 38 1
5 8 50 3
6 19 29 1
7 22 51 4
8 50 50
I do not answer questions offlist.
On Apr 28, 2009, at 2:56 AM, mathallan wrote:
Thanks for the answer David
Sum er the "sum insured" the maximal loss of the company. Claims, is
the
actually claim size. Group is wich type of business is insured.
Can you help me to solve the problem?
It
Thanks for the answer David
Sum er the "sum insured" the maximal loss of the company. Claims, is the
actually claim size. Group is wich type of business is insured.
Can you help me to solve the problem?
It is very difficult to determine "rightness" since you have omitted
essential backgroun
On Apr 27, 2009, at 5:19 PM, mathallan wrote:
I have to fit a generalized linear model in R, and I have never done
this
before, so I'm in very much doubt.
I have a dataset (of 4036 observations)
claims sum grp
1 3852 345702931
2
I have to fit a generalized linear model in R, and I have never done this
before, so I'm in very much doubt.
I have a dataset (of 4036 observations)
claims sum grp
1 3852 345702931
2 1194 7776468 1
3
15 matches
Mail list logo
