Re: [R] ANOVA in R

2006-09-14 Thread Russell Compton
Andrew, Peter,

Thanks both for the help, that's exactly what I was after. 

It is for a one-way ANOVA, looking at identifying differentially expressed
genes across time in a microarray dataset. Also, one of the datasets I'm
working with is unbalanced, so that additional code will be most useful.

Thanks again,

Russell Compton

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Peter
Dalgaard
Sent: 14 September 2006 10:26
To: Andrew Robinson
Cc: Russell Compton; [email protected]
Subject: Re: [R] ANOVA in R

Andrew Robinson <[EMAIL PROTECTED]> writes:

> Try
> 
> test <- data.frame(day.1=c(2,3,3,6,1), 
>  day.4=c(7,2,4,6,3),
>  day.8=c(2,8,7,8,4))
> 
> test
> 
> test.long <- reshape(test, direction="long", 
> varying=c("day.1","day.4","day.8"),
> v.names="response",
> timevar="day",
> times=names(test))
> 
> test.long$day <- factor(test.long$day)
> 
> test.long
> 
> aov(response ~ day, data=test.long)


Was a one-way ANOVA intended? He never said.

On a more elementary level,

y <- with(test, c(day.1,day.4,day.8))
day <- factor(rep(c(1,4,8),each=5)) # or gl(3,5,labels=c(1,4,8))
sub <- factor(rep(1:5,3))   # or gl(5,1,15)

print(data.frame(y,day,sub)) # just to show the point

anova(lm(y~day))# 1-way
anova(lm(y~day+sub))# 2-way

# This could be better for unbalanced designs:

drop1(lm(y~day+sub),test="F")

 

 
> 
> I hope that this helps,
> 
> Andrew
> 
> 
> On Thu, Sep 14, 2006 at 09:23:13AM +0100, Russell Compton wrote:
> > Despite having used R on a daily basis for the past two years, I'm
> > encountering some difficulty performing an ANOVA on my data. What I'm
trying
> > to do is the following:
> > 
> >  
> > 
> > Given data such as:
> > 
> >  
> > 
> > Day 1Day 4Day 8
> > 
> > 2  7  2
> > 
> > 3  2  8  
> > 
> > 3  4  7
> > 
> > 6  6  8
> > 
> > 1  3  4
> > 
> >  
> > 
> > I want to use ANOVA to determine if there is a significant change over
the
> > three days. In other stats packages I have used, I can just select this
data
> > and run the ANOVA function and get the F and p values. However in R, the
> > anova function seems to only work with a fitted model, eg. Linear
> > regression. This function seems to assume there is a relationship such
as
> > day1~ day 4 + day 8, but in my case there isn't - I just want to perform
an
> > ANOVA without regression. If anyone could point me in the right
direction
> > I'd greatly appreciate it,
> > 
> >  
> > 
> > Thanks
> > 
> > 
> > [[alternative HTML version deleted]]
> > 
> > __
> > [email protected] 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.
> 
> -- 
> Andrew Robinson  
> Department of Mathematics and StatisticsTel: +61-3-8344-9763
> University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599
> Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au
> 
> __
> [email protected] 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.
> 

-- 
   O__   Peter Dalgaard Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark  Ph:  (+45) 35327918
~~ - ([EMAIL PROTECTED])  FAX: (+45) 35327907

__
[email protected] 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.


Re: [R] ANOVA in R

2006-09-14 Thread Peter Dalgaard
Andrew Robinson <[EMAIL PROTECTED]> writes:

> Try
> 
> test <- data.frame(day.1=c(2,3,3,6,1), 
>  day.4=c(7,2,4,6,3),
>  day.8=c(2,8,7,8,4))
> 
> test
> 
> test.long <- reshape(test, direction="long", 
> varying=c("day.1","day.4","day.8"),
> v.names="response",
> timevar="day",
> times=names(test))
> 
> test.long$day <- factor(test.long$day)
> 
> test.long
> 
> aov(response ~ day, data=test.long)


Was a one-way ANOVA intended? He never said.

On a more elementary level,

y <- with(test, c(day.1,day.4,day.8))
day <- factor(rep(c(1,4,8),each=5)) # or gl(3,5,labels=c(1,4,8))
sub <- factor(rep(1:5,3))   # or gl(5,1,15)

print(data.frame(y,day,sub)) # just to show the point

anova(lm(y~day))# 1-way
anova(lm(y~day+sub))# 2-way

# This could be better for unbalanced designs:

drop1(lm(y~day+sub),test="F")

 

 
> 
> I hope that this helps,
> 
> Andrew
> 
> 
> On Thu, Sep 14, 2006 at 09:23:13AM +0100, Russell Compton wrote:
> > Despite having used R on a daily basis for the past two years, I'm
> > encountering some difficulty performing an ANOVA on my data. What I'm trying
> > to do is the following:
> > 
> >  
> > 
> > Given data such as:
> > 
> >  
> > 
> > Day 1Day 4Day 8
> > 
> > 2  7  2
> > 
> > 3  2  8  
> > 
> > 3  4  7
> > 
> > 6  6  8
> > 
> > 1  3  4
> > 
> >  
> > 
> > I want to use ANOVA to determine if there is a significant change over the
> > three days. In other stats packages I have used, I can just select this data
> > and run the ANOVA function and get the F and p values. However in R, the
> > anova function seems to only work with a fitted model, eg. Linear
> > regression. This function seems to assume there is a relationship such as
> > day1~ day 4 + day 8, but in my case there isn't - I just want to perform an
> > ANOVA without regression. If anyone could point me in the right direction
> > I'd greatly appreciate it,
> > 
> >  
> > 
> > Thanks
> > 
> > 
> > [[alternative HTML version deleted]]
> > 
> > __
> > [email protected] 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.
> 
> -- 
> Andrew Robinson  
> Department of Mathematics and StatisticsTel: +61-3-8344-9763
> University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599
> Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au
> 
> __
> [email protected] 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.
> 

-- 
   O__   Peter Dalgaard Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark  Ph:  (+45) 35327918
~~ - ([EMAIL PROTECTED])  FAX: (+45) 35327907

__
[email protected] 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.


Re: [R] ANOVA in R

2006-09-14 Thread Andrew Robinson


Try

test <- data.frame(day.1=c(2,3,3,6,1), 
 day.4=c(7,2,4,6,3),
 day.8=c(2,8,7,8,4))

test

test.long <- reshape(test, direction="long", 
  varying=c("day.1","day.4","day.8"),
  v.names="response",
  timevar="day",
  times=names(test))

test.long$day <- factor(test.long$day)

test.long

aov(response ~ day, data=test.long)


I hope that this helps,

Andrew


On Thu, Sep 14, 2006 at 09:23:13AM +0100, Russell Compton wrote:
> Despite having used R on a daily basis for the past two years, I'm
> encountering some difficulty performing an ANOVA on my data. What I'm trying
> to do is the following:
> 
>  
> 
> Given data such as:
> 
>  
> 
> Day 1Day 4Day 8
> 
> 2  7  2
> 
> 3  2  8  
> 
> 3  4  7
> 
> 6  6  8
> 
> 1  3  4
> 
>  
> 
> I want to use ANOVA to determine if there is a significant change over the
> three days. In other stats packages I have used, I can just select this data
> and run the ANOVA function and get the F and p values. However in R, the
> anova function seems to only work with a fitted model, eg. Linear
> regression. This function seems to assume there is a relationship such as
> day1~ day 4 + day 8, but in my case there isn't - I just want to perform an
> ANOVA without regression. If anyone could point me in the right direction
> I'd greatly appreciate it,
> 
>  
> 
> Thanks
> 
> 
>   [[alternative HTML version deleted]]
> 
> __
> [email protected] 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.

-- 
Andrew Robinson  
Department of Mathematics and StatisticsTel: +61-3-8344-9763
University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599
Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au

__
[email protected] 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.


[R] ANOVA in R

2006-09-14 Thread Russell Compton
Despite having used R on a daily basis for the past two years, I'm
encountering some difficulty performing an ANOVA on my data. What I'm trying
to do is the following:

 

Given data such as:

 

Day 1Day 4Day 8

2  7  2

3  2  8  

3  4  7

6  6  8

1  3  4

 

I want to use ANOVA to determine if there is a significant change over the
three days. In other stats packages I have used, I can just select this data
and run the ANOVA function and get the F and p values. However in R, the
anova function seems to only work with a fitted model, eg. Linear
regression. This function seems to assume there is a relationship such as
day1~ day 4 + day 8, but in my case there isn't - I just want to perform an
ANOVA without regression. If anyone could point me in the right direction
I'd greatly appreciate it,

 

Thanks


[[alternative HTML version deleted]]

__
[email protected] 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.


[R] ANOVA in R

2006-09-14 Thread Russell Compton
Despite having used R on a daily basis for the past two years, I’m
encountering some difficulty performing an ANOVA on my data. What I’m trying
to do is the following:
 
Given data such as:
 
Day 1    Day 4    Day 8
2  7  2
3  2  8  
3  4  7
6  6  8
1  3  4
 
I want to use ANOVA to determine if there is a significant change over the
three days. In other stats packages I have used, I can just select this data
and run the ANOVA function and get the F and p values. However in R, the
anova function seems to only work with a fitted model, eg. Linear
regression. This function seems to assume there is a relationship such as
day1~ day 4 + day 8, but in my case there isn’t – I just want to perform an
ANOVA without regression. If anyone could point me in the right direction
I’d greatly appreciate it,
 
Thanks

__
[email protected] 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.