help

-----Original Message-----
From: Marc Schwartz [mailto:mschwartz at medanalytics.com]
Sent: Friday, February 21, 2003 8:38 AM
To: 'Thanjavur Bragadeesh'; r-help at stat.math.ethz.ch
Subject: RE: [R] help

-----Original Message-----
From: r-help-admin at stat.math.ethz.ch 
[mailto:r-help-admin at stat.math.ethz.ch] On Behalf Of Thanjavur 
Bragadeesh
Sent: Monday, February 10, 2003 4:08 PM
To: r-announce at stat.math.ethz.ch
Subject: [R] help



Hi,

help.I would like to plot the mean and standard deviation 
against a factor 
in the xaxis- in other words I want an errorplot. I also want 
to put the 
standard deviation like a "T" on top of a barplot.  How can do 
this in R

Bragadeesh

Look at the barplot2() function in the gregmisc package on CRAN.

A quick example using barplot2() and faked numbers:


library(gregmisc)
barplot2(c(5, 8, 4, 6, 9), plot.ci = TRUE, ci.l = c(5, 8, 4, 6, 9),
ci.u = c(5.5, 8.3, 4.25, 6.7, 10.5))


This will give you 5 vertical bars, with the upper bounds of the
confidence intervals plotted on top of the bars. Note that I have
specified that the lower bounds are the same as the bar values, which
results in no lower intervals being plotted.  If you want both upper
and lower bounds plotted, use appropriate values for the 'ci.l'
argument.

There are other options for axis configuration, grid lines and so
forth in barplot2() if you should need them.

Hope that helps,

Marc Schwartz