On 08/24/04 13:50, Paolo Tommasini wrote:>Hi my name is Paolo Tommasini does anyone know how to compute a "mode" >( most frequent element ) for a distribution ?try median(density(my.vector)) John John Wilknson [wilks at dial.pipex.com]
Kjetil Brinchmann Halvorsen
2004-Aug-26 12:49 UTC
[R] Re:how to compute a "mode" of a distribution
John Wilkinson wrote:>On 08/24/04 13:50, Paolo Tommasini wrote: > > > >>Hi my name is Paolo Tommasini does anyone know how to compute a "mode" >>( most frequent element ) for a distribution ? >> >> > >try > >median(density(my.vector)) > > >That cannot be right. Try: > test <- rnorm(1000) > d <- density(test) > median(d) Error in median(d) : need numeric data > str(d) List of 7 $ x : num [1:512] -3.64 -3.63 -3.62 -3.60 -3.59 ... $ y : num [1:512] 2.16e-05 2.60e-05 3.11e-05 3.71e-05 4.43e-05 ... $ bw : num 0.221 $ n : int 1000 $ call : language density(x = test) $ data.name: chr "test" $ has.na : logi FALSE - attr(*, "class")= chr "density" > i <- which.max(d$y) > d$x[i] [1] 0.05625893 Kjetil halvorsen>John > >John Wilknson > >[wilks at dial.pipex.com] > >______________________________________________ >R-help at stat.math.ethz.ch mailing list >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > > > >
you are quite right ,I forgot to insert $x, I should have written median(density(my.vector)$x) that gives a median result but it is still not the proper result. density(my.vector)$y) gives the probabilites y of each x value The correst answer would be the value of x corresponding to y=max(density(my.vector)$y) see summary(density(my.vector) of course a plot(density(my.vector)) would reveal the mode easily by inspection. John ----- Original Message ----- From: "Kjetil Brinchmann Halvorsen" <kjetil at acelerate.com> To: "John Wilkinson" <wilks at dial.pipex.com> Cc: "Paolo Tommasini" <paolo at directwave.com.br>; <r-help at stat.math.ethz.ch> Sent: Thursday, August 26, 2004 1:49 PM Subject: Re: [R] Re:how to compute a "mode" of a distribution> John Wilkinson wrote: > > >On 08/24/04 13:50, Paolo Tommasini wrote: > > > > > > > >>Hi my name is Paolo Tommasini does anyone know how to compute a "mode" > >>( most frequent element ) for a distribution ? > >> > >> > > > >try > > > >median(density(my.vector)) > > > > > > > That cannot be right. Try: > > > test <- rnorm(1000) > > d <- density(test) > > median(d) > Error in median(d) : need numeric data > > str(d) > List of 7 > $ x : num [1:512] -3.64 -3.63 -3.62 -3.60 -3.59 ... > $ y : num [1:512] 2.16e-05 2.60e-05 3.11e-05 3.71e-05 4.43e-05 ... > $ bw : num 0.221 > $ n : int 1000 > $ call : language density(x = test) > $ data.name: chr "test" > $ has.na : logi FALSE > - attr(*, "class")= chr "density" > > i <- which.max(d$y) > > d$x[i] > [1] 0.05625893 > > > Kjetil halvorsen > > > > >John > > > >John Wilknson > > > >[wilks at dial.pipex.com] > > > >______________________________________________ > >R-help at stat.math.ethz.ch mailing list > >https://stat.ethz.ch/mailman/listinfo/r-help > >PLEASE do read the posting guide!http://www.R-project.org/posting-guide.html> > > > > > > > > > >