[R] An averaged model based on Quartile and Mean of Quartiles

[Ulrik Stervbo]

Hi Chalabi,

Maybe you can use ddply to look over the intervals and add the means as new
colum

Best wishes
Ulrik

David Winsemius <dwinsemius at comcast.net> schrieb am Sa., 21. Mai 2016 17:16:

>
> > On May 21, 2016, at 7:41 AM, ch.elahe via R-help <r-help at r-project.org>
> wrote:
> >
> > Hi all,
> >
> > Here is my df and I want to make an averaged model of my variables based
> on Time like the following:
> >
> >
> >   $ Protocol   : Factor w/ 48 levels "DP FS QTSE SAG",..: 2 3 43 42
> >   $ Time       : num  182 185 189 234 186 ...
> >
> >   $ Systemtype   : Factor w/ 2 levels "Aera XJ","AERA XQ": 1 1 1
> >   $ ADJ        : Factor w/ 2 levels "auto","manu": 1 1
> >   $ BR         : int  384 384 384 384 512 384
> >   $ TF         : int  10 10 13 7 7 5 5
> > I split my df into quartiles for time,
> >
> >
> >   df$quant=findInterval(df$Time, quantile(df$Time),
> rightmost.closed=TRUE)
> > by the column df$quant I see quartiles 1 to 4 for time, next I want to
> create a new column for example for BR and if df$quant==1 then put mean of
> first quartile of BR in that,if df$quant==2 put mean of second quartile of
> BR in that and so on. I do the same for every numeric variable. Does anyone
> know how to that?
> >
>
> Perhaps, but untested in the absence of a reproducible example and
> choosing to instead use a name for the dataframe which is not also a
> function name:
>
> aggregate( dfrm[ , sapply( dfrm, inherits, "numeric")],   # logical
> indexing for `[, j]`
>            dfrm['quant']      # using "[" keeps it a list
>            mean)
>
> --
>
> David Winsemius
> Alameda, CA, USA
>
> ______________________________________________
> 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.
>

	[[alternative HTML version deleted]]