R help - Apr 2008 - Running regression (lm, lrm) 100+ times and saving the results as matrix

An undergraduate here, so do not hesitate to let me know if you feel that
I''m heading in a wrong direction.

 

I have a data frame containing panel data across 10 years (hence 120
months).  I want to be able to run regression separately for each month (or
year).  The below shows how I ran the regression for each month, but I need
to know how I would combine the regression results together into a matrix
possibly.

 

Thank you!

 

Mike

 

-------------------------------

m = max(data$TIME)

# define regmatrix

 

for(i in 1:m){

 g=runreg(data, i)

 # attach g to regmatrix

}

 

runreg = function(data, index){

datainterim = subset(data, TIME==index)

g = lrm(X ~ A + B, datainterim)

return(g)

}


	[[alternative HTML version deleted]]

On 4/29/2008 7:27 AM, Mike H. Ryu wrote:
> An undergraduate here, so do not hesitate to let me know if you feel that
> I'm heading in a wrong direction.
> 
> I have a data frame containing panel data across 10 years (hence 120
> months).  I want to be able to run regression separately for each month (or
> year).  The below shows how I ran the regression for each month, but I need
> to know how I would combine the regression results together into a matrix
> possibly.
> 
> Thank you!
If by "results" you mean the coefficients, how about something like
this?

t(sapply(split(mydata, mydata$TIME),
     function(x){coef(lm(X ~ A + B, data = x))}))
> Mike
> 
> -------------------------------
> 
> m = max(data$TIME)
> 
> # define regmatrix
> 
> for(i in 1:m){
> 
>  g=runreg(data, i)
> 
>  # attach g to regmatrix
> 
> }
> 
> runreg = function(data, index){
> 
> datainterim = subset(data, TIME==index)
> 
> g = lrm(X ~ A + B, datainterim)
> 
> return(g)
> 
> }
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code. 
-- 
Chuck Cleland, Ph.D.
NDRI, Inc. (www.ndri.org)
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 512-0171 (M, W, F)
fax: (917) 438-0894

Much thanks to Chuck Cleland for the following solution:



t(sapply(split(mydf, mydf$TIME), function(x){coefficients(summary(glm(X ~ A
+ B, data = x, family=binomial)))}))

To see what's in the 12 columns of the matrix returned by those lines, look
at the results this way:

lapply(split(mydf, mydf$TIME), function(x){coefficients(summary(glm(X ~ A +
B, data = x, family=binomial)))})



-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On
Behalf Of Mike H. Ryu
Sent: Tuesday, April 29, 2008 7:27 AM
To: r-help at r-project.org
Subject: [R] Running regression (lm, lrm) 100+ times and saving the results
as matrix

An undergraduate here, so do not hesitate to let me know if you feel that
I'm heading in a wrong direction.

 

I have a data frame containing panel data across 10 years (hence 120
months).  I want to be able to run regression separately for each month (or
year).  The below shows how I ran the regression for each month, but I need
to know how I would combine the regression results together into a matrix
possibly.

 

Thank you!

 

Mike

 

-------------------------------

m = max(data$TIME)

# define regmatrix

 

for(i in 1:m){

 g=runreg(data, i)

 # attach g to regmatrix

}

 

runreg = function(data, index){

datainterim = subset(data, TIME==index)

g = lrm(X ~ A + B, datainterim)

return(g)

}


	[[alternative HTML version deleted]]

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.