The standard R boxplot appears to use quartiles to determine the height of the rectangles and a range parameter - RNG - (default=1.5 I think) that determines the length of the whiskers as <= RNG x Interquartile Range. Is it possible to instead define the range as extending to the 95th percentile? If so, how would this be done? nb, I'm plotting multiple boxplots on a single chart so a function that applies to multiple series would be useful. Thanks, Nick -------------- next part -------------- An HTML attachment was scrubbed... URL: https://stat.ethz.ch/pipermail/r-help/attachments/20011126/49c9a092/attachment.html
On Mon, Nov 26, 2001 at 11:25:36AM +1300, Nick Davis wrote:> The standard R boxplot appears to use quartiles to determine the height of > the rectangles and a range parameter - RNG - (default=1.5 I think) that > determines the length of the whiskers as <= RNG x Interquartile Range. Is > it possible to instead define the range as extending to the 95th percentile? > If so, how would this be done?You need to create the right kind of thing to call bxp -- which has some documentation but basically it is the z component that you need to construct. You can call boxplot with plot=FALSE to get it to do most of the work and then just fiddle with the parts you want to change> > nb, I'm plotting multiple boxplots on a single chart so a function that > applies to multiple series would be useful. > > Thanks, > > Nick-- +---------------------------------------------------------------------------+ | Robert Gentleman phone : (617) 632-5250 | | Associate Professor fax: (617) 632-2444 | | Department of Biostatistics office: M1B28 | Harvard School of Public Health email: rgentlem at jimmy.dfci.harvard.edu | +---------------------------------------------------------------------------+ -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Nick, The last things to be ready in my porting the Hmisc library to R are Trellis-related functions. When I have ported panel.bpplot, which extends work of Jeff Banfield, that panel function offers a huge number of options for constructing box plots. You can specify multiple percentiles and various ranges of raw data (rug plots) to show. If you are already familiar with Grid/Lattice and would like to port panel.bpplot as a replacement for panel.bwplot, I would be glad to send you the code. Frank Harrell On Mon, 26 Nov 2001 11:25:36 +1300 Nick Davis <Nick.Davis at treasury.govt.nz> wrote:> The standard R boxplot appears to use quartiles to determine the height of > the rectangles and a range parameter - RNG - (default=1.5 I think) that > determines the length of the whiskers as <= RNG x Interquartile Range. Is > it possible to instead define the range as extending to the 95th percentile? > If so, how would this be done? > > nb, I'm plotting multiple boxplots on a single chart so a function that > applies to multiple series would be useful. > > Thanks, > > Nick >-- Frank E Harrell Jr Prof. of Biostatistics & Statistics Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences U. Virginia School of Medicine http://hesweb1.med.virginia.edu/biostat -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._