R help - Aug 2016 - Looping through different groups of variables in models

Hi all,

I am having trouble wrapping my head around a probably simple issue:

After using the reshape package, I have a melted dataframe with the columns
group (factor), time (int), condition (factor), value(int).

These are experimental data. The data were obtained from different
treatment groups (group) under different conditions at different time
points.

I would now like to perform ANOVA, boxplots and calculate means to compare
groups for all combinations of conditions and time points with something
like

fit <- lm(value~group, data=[subset of data with combination of
condition/timepoint])
summary (fit)
p <- ggplot([subset of data with combination of condition/timepoint],
aes(x= group, y=value)) + geom_boxplot ()
print (p)
tapply ([subset of data with combination of condition/timepoint]$value,
subset of data with combination of condition/timepoint]$group, mean)

How can I loop through these combinations and output the data in an elegant
way?

Thanks so much!

Best,

Kai

	[[alternative HTML version deleted]]

Kai:

1. I think that this is a very bad idea, statistically, if I
understand you correctly. Generally, your model should incorporate all
groups, time points, and conditions together, not individually.

2. But plotting results in "small multiples" -- aka "trellis
plots"
may be useful. This is done in ggplot through "faceting" which you
could read up on and try (I use lattice, not ggplot, to do this sort
of thing, so can't help with code).

3. However, I think your question is mostly statistical in nature
(define "elegant"), and if so, is off topic here. You might therefore
try stats.stackexchange.com instead to get ideas on how to approach
your data, solicit other opinions on whether what you want to do makes
sense (and if not, what else), etc. Or, perhaps better yet, consult a
local statistical resource.

Cheers,
Bert


Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Wed, Aug 31, 2016 at 2:58 PM, Kai Mx <govokai at gmail.com>
wrote:
> Hi all,
>
> I am having trouble wrapping my head around a probably simple issue:
>
> After using the reshape package, I have a melted dataframe with the columns
> group (factor), time (int), condition (factor), value(int).
>
> These are experimental data. The data were obtained from different
> treatment groups (group) under different conditions at different time
> points.
>
> I would now like to perform ANOVA, boxplots and calculate means to compare
> groups for all combinations of conditions and time points with something
> like
>
> fit <- lm(value~group, data=[subset of data with combination of
> condition/timepoint])
> summary (fit)
> p <- ggplot([subset of data with combination of condition/timepoint],
> aes(x= group, y=value)) + geom_boxplot ()
> print (p)
> tapply ([subset of data with combination of condition/timepoint]$value,
> subset of data with combination of condition/timepoint]$group, mean)
>
> How can I loop through these combinations and output the data in an elegant
> way?
>
> Thanks so much!
>
> Best,
>
> Kai
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.

Hi Kai,
Perhaps something like this:

kmdf<-data.frame(group=rep(c("exp","cont"),each=50),
 time=factor(rep(1:5,20)),
 condition=rep(rep(c("hot","cold"),each=25),2),
 value=sample(100:200,100))
for(timeindx in levels(kmdf$time)) {
 for(condindx in levels(kmdf$condition)) {
  cat("Time",timeindx,"Condition",condindx,"\n")
  subdat<-kmdf[kmdf$time == timeindx & kmdf$condition == condindx,]
  fit<-lm(value~group,subdat)
  print(summary(fit))
  plot(subdat$group,subdat$value)
  by(subdat$value,subdat$group,mean)
 }
}

Getting elegant output is another matter. Have a look at packages
meant to produce fancier R output.

Jim


On Thu, Sep 1, 2016 at 7:58 AM, Kai Mx <govokai at gmail.com>
wrote:
> Hi all,
>
> I am having trouble wrapping my head around a probably simple issue:
>
> After using the reshape package, I have a melted dataframe with the columns
> group (factor), time (int), condition (factor), value(int).
>
> These are experimental data. The data were obtained from different
> treatment groups (group) under different conditions at different time
> points.
>
> I would now like to perform ANOVA, boxplots and calculate means to compare
> groups for all combinations of conditions and time points with something
> like
>
> fit <- lm(value~group, data=[subset of data with combination of
> condition/timepoint])
> summary (fit)
> p <- ggplot([subset of data with combination of condition/timepoint],
> aes(x= group, y=value)) + geom_boxplot ()
> print (p)
> tapply ([subset of data with combination of condition/timepoint]$value,
> subset of data with combination of condition/timepoint]$group, mean)
>
> How can I loop through these combinations and output the data in an elegant
> way?
>
> Thanks so much!
>
> Best,
>
> Kai
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.