R help - Feb 2009 - Syntax in taking log to transfrom the data to fit Gaussian distribution

Hi,

I have a data set (weight) that does not follow the Gaussian (Normal)
distribution. However, I have to transform the data before applying the
Gaussian distribution. I used this syntax and used log(weight) as:
posJy.model<-glm(log(weight) ~ factor(pos),
family=gaussian(link='identity'), subset=Soil=="Jy"). This
syntax COULD NOT
transform the data. But if I transform the data before applying the model,
then it works. The syntax is: 

weight<-log(weight)
posJy.model<-glm(weight ~ factor(pos),
family=gaussian(link='identity'),
subset=Soil=="Jy")


What is wrong with the first syntax. I have seen many scripts that the R
users have applied any function (log, -, ...) directly inside the
formulation. 

It will help me to understand the grammar of R.

Thank you.
Saeed




-- 
View this message in context:
http://www.nabble.com/Syntax-in-taking-log-to-transfrom-the-data-to-fit-Gaussian-distribution-tp22179656p22179656.html
Sent from the R help mailing list archive at Nabble.com.

Petr Pikal
petr.pikal at precheza.cz
724008364, 581252140, 581252257


r-help-bounces at r-project.org napsal dne 24.02.2009 11:46:07:
> 
> Hi,
> 
> I have a data set (weight) that does not follow the Gaussian (Normal)
> distribution. However, I have to transform the data before applying the
> Gaussian distribution. I used this syntax and used log(weight) as:
> posJy.model<-glm(log(weight) ~ factor(pos),
> family=gaussian(link='identity'), subset=Soil=="Jy").
This syntax COULD
NOT
> transform the data. But if I transform the data before applying the 
model,

Without going too much into details, there is nothing wrong in log inside 
formula
try

summary(glm(lot1 ~ log(u), data=clotting, family=Gamma))
summary(glm(log(lot1) ~ log(u), data=clotting, family=Gamma))
> then it works. The syntax is: 
> 
> weight<-log(weight)
> posJy.model<-glm(weight ~ factor(pos),
family=gaussian(link='identity'),
> subset=Soil=="Jy")
> 
> 
> What is wrong with the first syntax. 
AFAIK, nothing

Regards
Petr

I have seen many scripts that the R
> users have applied any function (log, -, ...) directly inside the
> formulation. 
> 
> It will help me to understand the grammar of R.
> 
> Thank you.
> Saeed
> 
> 
> 
> 
> -- 
> View this message in context: 
http://www.nabble.com/Syntax-in-taking-log-to-
> transfrom-the-data-to-fit-Gaussian-distribution-tp22179656p22179656.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
> PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

On Tue, Feb 24, 2009 at 02:46:07AM -0800, Saeed Ahmadi
wrote:
> I have a data set (weight) that does not follow the Gaussian (Normal)
> distribution. However, I have to transform the data before applying the
> Gaussian distribution. I used this syntax and used log(weight) as:
> posJy.model<-glm(log(weight) ~ factor(pos),
> family=gaussian(link='identity'), subset=Soil=="Jy").
This syntax COULD NOT
> transform the data. But if I transform the data before applying the model,
> then it works. The syntax is: 
> 
> weight<-log(weight)
> posJy.model<-glm(weight ~ factor(pos),
family=gaussian(link='identity'),
> subset=Soil=="Jy")
That should work. Maybe you should tell us what happened or why you
think it didn't work. Error messages? Unexpected results?

cu
	Philipp

-- 
Dr. Philipp Pagel
Lehrstuhl f?r Genomorientierte Bioinformatik
Technische Universit?t M?nchen
Wissenschaftszentrum Weihenstephan
85350 Freising, Germany
http://mips.gsf.de/staff/pagel