Same regression per sub-group: apply?

Dear helpers,

I've come up with what is probably a simple problem, but I cannot
find the solution. I have a data-set containing survey-data from
several countries. What I want to do is to perform some regression
analyses, for each country separately. The question is, how to do
this nicely (thus without repeating the same syntax with another
`subset' argument).

I thought of the following:

model.per.country <- tapply(data, data$COUNTRY, function(x) glm
(dependent.var ~ FEMALE + AGE + EDUCLIN + (), family=binomial,
data=capital))

But this does not work. What goes wrong, I think, is that the
dependent variable is clustered according to `Country', but not so
for the predictors. The error message I received:

Error in tapply(dat, dat$COUNTRY, function(x) glm(participate ~
FEMALE +  :
arguments must have same length

Could you please help me solving this little problem?

Thanks in advance:

Rense Nieuwenhuis



[[alternative HTML version deleted]]

try something like this (untested):

dataCountry <- split(data, data$COUNTRY)
model.per.country <- lapply(dataCountry, function (x) {
    glm(dependent.var ~ FEMALE + AGE + EDUCLIN, family = binomial, 
data = x)
})


I hope it helps.

Best,
Dimitris

----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://med.kuleuven.be/biostat/
     http://www.student.kuleuven.be/~m0390867/dimitris.htm


----- Original Message ----- 
From: "Rense Nieuwenhuis" <r.nieuwenhuis at student.ru.nl>
To: <r-help at stat.math.ethz.ch>
Sent: Friday, December 07, 2007 9:32 AM
Subject: [R] Same regression per sub-group: apply?


  

Dear helpers,

I've come up with what is probably a simple problem, but I cannot
find the solution. I have a data-set containing survey-data from
several countries. What I want to do is to perform some regression
analyses, for each country separately. The question is, how to do
this nicely (thus without repeating the same syntax with another
`subset' argument).

I thought of the following:

model.per.country <- tapply(data, data$COUNTRY, function(x) glm
(dependent.var ~ FEMALE + AGE + EDUCLIN + (), family=binomial,
data=capital))

But this does not work. What goes wrong, I think, is that the
dependent variable is clustered according to `Country', but not so
for the predictors. The error message I received:

Error in tapply(dat, dat$COUNTRY, function(x) glm(participate ~
FEMALE +  :
arguments must have same length
    
Could you please help me solving this little problem?

Thanks in advance:

Rense Nieuwenhuis



[[alternative HTML version deleted]]

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

Dear helpers,

I've come up with what is probably a simple problem, but I cannot
find the solution. I have a data-set containing survey-data from
several countries. What I want to do is to perform some regression
analyses, for each country separately. The question is, how to do
this nicely (thus without repeating the same syntax with another
`subset' argument).

I thought of the following:

model.per.country <- tapply(data, data$COUNTRY, function(x) glm
(dependent.var ~ FEMALE + AGE + EDUCLIN + (), family=binomial,
data=capital))

But this does not work. What goes wrong, I think, is that the
dependent variable is clustered according to `Country', but not so
for the predictors. The error message I received:

Error in tapply(dat, dat$COUNTRY, function(x) glm(participate ~
FEMALE +  :
       arguments must have same length

 >

Could you please help me solving this little problem?

Thanks in advance:

Rense Nieuwenhuis



       [[alternative HTML version deleted]]