Full_Name: Michael Toews Version: 2.1.1 OS: WinXP (SP2) Submission from: (NULL) (142.58.206.114) This is a simple feature request concerning the display of the axis in the coplot function. Consider an arbritrary set of values from 1990-1999: date <- as.Date(paste(rep(1990:1999,each=365),1:365),"%Y %j") mon <- factor(months(date,T),levels=month.abb) value <- runif(3650) # Now create coplot for each month: coplot(value ~ date | mon) # The resulting date on the x-axis is not like the axis produced from: plot(value ~ date)
coplot using Date object (PR#8147)
2 messages · mwtoews at sfu.ca, Brian Ripley
On Tue, 20 Sep 2005 mwtoews at sfu.ca wrote:
Full_Name: Michael Toews Version: 2.1.1 OS: WinXP (SP2) Submission from: (NULL) (142.58.206.114) This is a simple feature request concerning the display of the axis in the coplot function. Consider an arbritrary set of values from 1990-1999: date <- as.Date(paste(rep(1990:1999,each=365),1:365),"%Y %j") mon <- factor(months(date,T),levels=month.abb) value <- runif(3650) # Now create coplot for each month: coplot(value ~ date | mon) # The resulting date on the x-axis is not like the axis produced from: plot(value ~ date)
Why do you expect it to be? plot() is generic, and coplot() is not: plot(value ~ date) is just a round-the-houses way to generate the call plot(date, value) (and that is why it happens to work with Date objects). Think of coplot() as a limited predecessor of lattice graphics, which also precedes class "Date". Lattice currently handles POSIXct date-times, and it would be nice if it handled Date too.
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595