R help - Jul 2011 - Add a density line to a cumulative histogram - second try

Hi list,

this is my second try for first post on this list (I tried to post via email
and nothing appeared in my email-inbox, so now I try to use the
nabble-web-interface) - I hope that you will only have to read one post in
your inbox! Okay, my question ...

I was able to plot a histogram and add the density()-line to this plot.
I was able to plot a cumulative form of this histogram.
Yet, I was not able to add the density line to this cumulative histogram.

You can watch a picture of the histograms here:
http://www.jochen-bauer.net/downloads/histo-cumulative-density.png

Source:

# Histogramm
histo <- hist( hgd$V1, freq=FALSE )

# Dichte-Sch?tzer-Funktion reinzeichnen
denspoints <- density( hgd$V1 )
lines( denspoints, col="GREEN", lwd=2 )

# Kumulative Verteilung relativer H?ufigkeiten
relcum <- cumsum( histo$counts ) / sum(histo$counts)
barplot( relcum, names.arg=round( histo$mids, 2), col="green",
ylab="Wahrscheinlichkeit")

# Kumulative Dichtefunktion ?

Thanks in advance - Jochen

--
View this message in context:
http://r.789695.n4.nabble.com/Add-a-density-line-to-a-cumulative-histogram-second-try-tp3666969p3666969.html
Sent from the R help mailing list archive at Nabble.com.

This is a hack which uses the output of the density function.

rrr <- cumsum(diff(cumsum(denspoints$y))*diff(denspoints$x))
lines(denspoints$x[-512],rrr)

I don't believe this is the best solution, because it is not a direct
estimate of the cumulative density function.  I think there are methods for
a direct estimate, and the name James Ramsay comes to mind.  You may want to
google his work on estimating strictly monotone functions.

Hope that helps,
-tgs


On Thu, Jul 14, 2011 at 5:03 AM, Jochen1980 <info@jochen-bauer.net> wrote:
> Hi list,
>
> this is my second try for first post on this list (I tried to post via
> email
> and nothing appeared in my email-inbox, so now I try to use the
> nabble-web-interface) - I hope that you will only have to read one post in
> your inbox! Okay, my question ...
>
> I was able to plot a histogram and add the density()-line to this plot.
> I was able to plot a cumulative form of this histogram.
> Yet, I was not able to add the density line to this cumulative histogram.
>
> You can watch a picture of the histograms here:
> http://www.jochen-bauer.net/downloads/histo-cumulative-density.png
>
> Source:
>
> # Histogramm
> histo <- hist( hgd$V1, freq=FALSE )
>
> # Dichte-Schätzer-Funktion reinzeichnen
> denspoints <- density( hgd$V1 )
> lines( denspoints, col="GREEN", lwd=2 )
>
> # Kumulative Verteilung relativer Häufigkeiten
> relcum <- cumsum( histo$counts ) / sum(histo$counts)
> barplot( relcum, names.arg=round( histo$mids, 2), col="green",
> ylab="Wahrscheinlichkeit")
>
> # Kumulative Dichtefunktion ?
>
> Thanks in advance - Jochen
>
> --
> View this message in context:
>
http://r.789695.n4.nabble.com/Add-a-density-line-to-a-cumulative-histogram-second-try-tp3666969p3666969.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@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.
>
	[[alternative HTML version deleted]]

Thanks, I found the function ecdf() which does the job.

plot( ecdf( nvtpoints), col="BLUE", lwd=1, add=TRUE )


--
View this message in context:
http://r.789695.n4.nabble.com/Add-a-density-line-to-a-cumulative-histogram-second-try-tp3666969p3669310.html
Sent from the R help mailing list archive at Nabble.com.