How would I color points conditional on their value in a plot of a time series

Excellent! Worked like a charm. Thanks.

--Chris Ryan

On Tue, May 1, 2018 at 4:33 PM, William Dunlap <wdunlap at tibco.com> wrote:

The ts method for plot() is quirky.  You can use the default method:

 plot(as.vector(time(ttt)), as.vector(ttt), type = "p", col=ifelse(ttt<8,
"black", "red"))


Bill Dunlap
TIBCO Software
wdunlap tibco.com

On Tue, May 1, 2018 at 1:17 PM, Christopher W Ryan <cryan at binghamton.edu>
wrote:

How would I color points conditional on their value in a plot of a time
series.  Something like this:

## demonstration data
ttt <- ts(rpois(12, lambda = 8), start = c(2000, 1), freq = 4)
ttt
plot(ttt, type = "p")

## doesn't work--all points the same color
plot(ttt, type = "p", col = ifelse(ttt < 8, "black", "red"))

## also doesn't work--all points the same color
q <- as.numeric(ttt)
q
plot(ttt, type = "p", col = ifelse(q < 8, "black", "red"))


## works OK with a simple, non-time-series scatterplot, as in

sss <- data.frame(x = rpois(12, lambda = 8), y = rnorm(12, mean = 100, sd
=
25))
with(sss, plot(y ~ x, col = ifelse(y > 100, "black", "red")))

## but I am missing something about time series.

Thanks.

--Chris Ryan
Broome County Health Department
and Binghamton University
Binghamton, NY

        [[alternative HTML version deleted]]

	[[alternative HTML version deleted]]