I really want to do this:
abline(
a=tan(-kT*pi/180),
b=kY-tan(-kT*pi/180)*kX
)
where kX,kY and kT are vectors of equal length. But I can't do that
with abline unless I use a loop, and I haven't figured out the least
unelegant way of writing the loop yet. So is there a way to do this
without a loop?
Or if I am to resort to the loop, what's the best way of doing it
considering that I have some missing data? Here's the mess that I
wrote.
converge <- na.omit(data.frame(kX,kY,kT))
for (z in (length(converge$kT)))
{abline(
a=tan(converge$kT[z]*pi/180),
b=converge$kY[z]-tan(-converge$kT[z]*converge$kX[z]*pi/180)
)}
I think the missing data are causing the problem; this happens when I run:
Error in int_abline(a = a, b = b, h = h, v = v, untf = untf, ...) :
'a' and 'b' must be finite
On 2/04/2009, at 7:04 AM, Thomas Levine wrote:> I really want to do this: > > abline( > a=tan(-kT*pi/180), > b=kY-tan(-kT*pi/180)*kX > ) > > where kX,kY and kT are vectors of equal length. But I can't do that > with abline unless I use a loop, and I haven't figured out the least > unelegant way of writing the loop yet. So is there a way to do this > without a loop? > > Or if I am to resort to the loop, what's the best way of doing it > considering that I have some missing data? Here's the mess that I > wrote. > > converge <- na.omit(data.frame(kX,kY,kT)) > for (z in (length(converge$kT))) > {abline( > a=tan(converge$kT[z]*pi/180), > b=converge$kY[z]-tan(-converge$kT[z]*converge$kX[z]*pi/180) > )} > > I think the missing data are causing the problem; this happens when > I run: > > Error in int_abline(a = a, b = b, h = h, v = v, untf = untf, ...) : > 'a' and 'b' must be finiteThe help for abline explicitly states that a and b must be ``single values''; so no vectorization appears to be possible, as abline is currently written. Hence you are stuck with a for-loop. There appears to be nothing wrong with the for-loop that you've written, at first blush at least. There won't be NAs in ``converge'' since you've very cleverly used na.omit. So ``missing data'' are NOT the problem. The problem is then (probably) that some of your data are yielding infinite values of tan(). ***Look*** at the values in converge. ***Look*** at the values of a and b produced in your loop and see where you're getting infinite values. cheers, Rolf Turner ###################################################################### Attention:\ This e-mail message is privileged and confid...{{dropped:9}}
May be:
plot(c(-1, 1) , c(-1, 1), type = "n")
n = 4
a = rep(0, n)
b = 1:n/n
fun = function(i, a, b, col = 1 , ...) {
abline(a[i], b[i], col = col[i], ...)
}
lapply(1:n, fun, a=a, b=b, col = 1:n)
Andrea
Thomas Levine wrote:> I really want to do this:
>
> abline(
> a=tan(-kT*pi/180),
> b=kY-tan(-kT*pi/180)*kX
> )
>
> where kX,kY and kT are vectors of equal length. But I can't do that
> with abline unless I use a loop, and I haven't figured out the least
> unelegant way of writing the loop yet. So is there a way to do this
> without a loop?
>
> Or if I am to resort to the loop, what's the best way of doing it
> considering that I have some missing data? Here's the mess that I
> wrote.
>
> converge <- na.omit(data.frame(kX,kY,kT))
> for (z in (length(converge$kT)))
> {abline(
> a=tan(converge$kT[z]*pi/180),
> b=converge$kY[z]-tan(-converge$kT[z]*converge$kX[z]*pi/180)
> )}
>
> I think the missing data are causing the problem; this happens when I run:
>
> Error in int_abline(a = a, b = b, h = h, v = v, untf = untf, ...) :
> 'a' and 'b' must be finite
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>