day coefficient se
Monday 1.2 0.2
Tuesday etc.
plot(day,coefficient)
segments(day,coefficient+se,day,coefficient-se)
I get the error bars plotted but they are not on the correct y-scale
,as I expected, being clustered at the left-hand edge of the plot.
Second, I am modelling a time trend in a model formula using a
spline. So I do the following :
time <- 1:730
fit <- glm(outcome ~ ns(time,df=12),family=poisson)
plot.gam(fit)
The y-scale of the plot naturally runs from 1 to 730 but I would like
this to be in dates from 04/01/93 to 03/31/95. I have tried adding
dates to the model formula but S then tried unsucessfully to plot
a persp plot.
I know there will be simple answers to these points.
Sorry if seems trivial but I always find controlling graphics the
most difficult part of S to understand.
yours sincerely
Andrew McCulloch
London School of Hygiene and Tropical Medicine
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news