<57d3c6e9-402b-4991-a53b-71e25af78...@comcast.net> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0
No worries there =3B-)) But thanks for the reminder all the same! =20 Monica ---------------------------------------- > CC: pisican...@hotmail.com=3b r-h...@stat.math.ethz.ch > From: dwinsem...@comcast.net > To: r.m.k...@gmail.com > Subject: Re: [R] intersection of 2 density curves > Date: Thu=2C 8 Oct 2009 11:16:22 -0400 > > It should also be remembered that there will often be multiple > intersections of such density estimates. > > I hope this is not being done in support of a data-driven > discretization of two group comparisons. Such practices are to be > eschewed (as they are even worse than obfuscation). > > -- > David Winsemius. > > > On Oct 8=2C 2009=2C at 10:01 AM=2C Rainer M Krug wrote: > >> On Thu=2C Oct 8=2C 2009 at 3:48 PM=2C Monica Pisica >> wrote: >> >>> >>> >>> Hi=2C >>> >>> I would like to find out the coordinates of the intersection points >>> of 2 >>> density curves. I did a search but i didn't get any significant >>> results. I >>> really hope some of you have some ideas. here it is an example: >>> >>> set.seed(123) >>> x1 <- rnorm(100=2C 1=2C 1) >>> x2 <- rnorm(100=2C 0=2C 1) >>> d1 <- density(x1) >>> d2 <- density(x2) >>> >>> plot(d1=2C col =3D 2) >>> lines(d2=2C col =3D 4) >>> >>> >> if you specify n=2C from ant to in density() and keep them the same=2C >> then the >> density estimates are done for the same x-coordinates. Therefore you >> can >> compare d1$y with d2$y to deternine=2C where they intersect: >> >> d2 <- density(x2=2C from=3D-3=2C to=3D5=2C n=3D2048) >> d1 <- density(x1=2C from=3D-3=2C to=3D5=2C n=3D2048) >> plot(d1=2C col =3D 2) >> lines(d2=2C col =3D 4) >> d1$y < d2$y >> intersX <- d1$x[as.logical(abs(diff(d1$y < d2$y)))] >> intersY <- d1$y[as.logical(abs(diff(d1$y < d2$y)))] >> abline(v=3Dinters) >> abline(h=3DintersY) >> >> Cheers=2C >> >> Rainer >> >> >>> Now i would like to know the coordinates of the intersection point. >>> I did >>> some "trial and error" and i come up with these .... (that means i >>> did >>> several vertical and horizontal lines through the graph until >>> visually i >>> "hit" the intersection point) >>> >>> abline(v =3D 0.35=2C lty =3D 2) >>> abline(h =3D 0.34=2C lty =3D 2) >>> >>> I really appreciate any hints you may have=2C thanks so much=2C >>> >>> Monica >>> _________________________________________________________________ >>> http://clk.atdmt.com/GBL/go/171222985/direct/01/ >>> ______________________________________________ >>> 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=2C minimal=2C self-contained=2C reproducible code= . >>> >> >> >> >> -- >> Rainer M. Krug=2C PhD (Conservation Ecology=2C SUN)=2C MSc (Conservation >> Biology=2C >> UCT)=2C Dipl. Phys. (Germany) >> >> Centre of Excellence for Invasion Biology >> Natural Sciences Building >> Office Suite 2039 >> Stellenbosch University >> Main Campus=2C Merriman Avenue >> Stellenbosch >> South Africa >> >> Cell: +27 - (0)83 9479 042 >> Fax: +27 - (0)86 516 2782 >> Fax: +49 - (0)721 151 334 888 >> email: rai...@krugs.de >> >> Skype: RMkrug >> Google: r.m.k...@gmail.com >> >> [[alternative HTML version deleted]] >> >> ______________________________________________ >> 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=2C minimal=2C self-contained=2C reproducible code. > > David Winsemius=2C MD > Heritage Laboratories > West Hartford=2C CT > =0A= _________________________________________________________________=0A= Hotmail: Powerful Free email with security by Microsoft.=0A= ______________________________________________ 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.
