Re: [R] densities greater than 1 for values within an (0, 1) intervall
How does one then interpret kernel density distributions with values greater than one? My output from the density function. --- density(delt.m[[1]][,6], na.rm=TRUE) Call: density.default(x = delt.m[[1]][, 6], na.rm = TRUE) Data: delt.m[[1]][, 6] (171 obs.); Bandwidth 'bw' = 0.004501 x y Min. :-0.05211 Min. : 0.00586 1st Qu.:-0.02177 1st Qu.: 0.43632 Median : 0.00856 Median : 3.08833 Mean : 0.00856 Mean : 8.23366 3rd Qu.: 0.03889 3rd Qu.:14.97542 Max. : 0.06923 Max. :30.04107 -- View this message in context: http://r.789695.n4.nabble.com/densities-greater-than-1-for-values-within-an-0-1-intervall-tp2286439p3228268.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.
Re: [R] densities greater than 1 for values within an (0, 1) intervall
The same way you interpret densities less than one? On Jan 20, 2011, at 2:28 PM, Paul Ramer wrote: How does one then interpret kernel density distributions with values greater than one? The same way you interpret densities less than one? density != probability My output from the density function. --- density(delt.m[[1]][,6], na.rm=TRUE) Call: density.default(x = delt.m[[1]][, 6], na.rm = TRUE) Data: delt.m[[1]][, 6] (171 obs.); Bandwidth 'bw' = 0.004501 x y Min. :-0.05211 Min. : 0.00586 1st Qu.:-0.02177 1st Qu.: 0.43632 Median : 0.00856 Median : 3.08833 Mean : 0.00856 Mean : 8.23366 3rd Qu.: 0.03889 3rd Qu.:14.97542 Max. : 0.06923 Max. :30.04107 -- David Winsemius, MD West Hartford, CT __ 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.
Re: [R] densities greater than 1 for values within an (0, 1) intervall
There is no constraint on the magnitude of probability density values, though the area under the curve must be equal to one. You may be thinking of cumulative probability distributions? If so, take a look at smoothed.df() in library (cwhmisc). Katja Hillmann katja.hillm...@wiso.uni-hamburg.de wrote: Hello, I used the command kdensity in order to calculate the density of fractions/ratios (e.g. number of longterm unemployed on total unemployment). Thus I try to calculate the denisty of values less than 1. However, the values of the kernel densitiy R provided (y-scale) are all greater than 1. Where is the problem and how may I solve it? Does R have problems in calculating distributions of variables within an intervall of 0 and 1? Best, Katja __ 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. --- Jeff NewmillerThe . . Go Live... DCN:jdnew...@dcn.davis.ca.usBasics: ##.#. ##.#. Live Go... Live: OO#.. Dead: OO#.. Playing Research Engineer (Solar/BatteriesO.O#. #.O#. with /Software/Embedded Controllers) .OO#. .OO#. rocks...1k --- Sent from my phone. Please excuse my brevity. __ 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.