[R] The standard deviation of measurement 1 with respect to measurement 2
Mark Lyman
mark.lyman at atk.com
Wed Aug 13 23:02:33 CEST 2008
Firas Swidan <frsswdn <at> gmail.com> writes:
>
> Hi,
>
> I have two (different types of) measurements, say X and Y, resulting from
> the same set of experiments. So X and Y are paired: (x_1, y_1), (x_2, y_2),
> ...
>
> I am trying to calculate the standard deviation of Y with respect to X. In
> other words, in terms of the scatter plot of X and Y, I would like to divide
> it into bins along the X-axis and for each bin calculate the standard
> deviation along the Y results in that bin. (Though I am not totally sure,
> this seems to remind me of the conditional expectation of Y given X - maybe
> it is called the conditional deviation?)
>
> Is their a built in procedure in R for calculating the above? Otherwise,
> what would be the easiest way to achieve it? (factors maybe?)
>
> Thankful for the help,
> Firas.
>
Something like the following should give you what you want:
> x <- rnorm(50)
> y <- rnorm(50)
> tapply(y, cut(x, 10, include.lowest=TRUE), sd)
[-2.19,-1.75] (-1.75,-1.3] (-1.3,-0.86] (-0.86,-0.415] (-0.415,0.029]
0.7569111 0.1671267 0.5620591 1.1280510 0.7772356
(0.029,0.473] (0.473,0.918] (0.918,1.36] (1.36,1.81] (1.81,2.25]
0.5600363 0.7681090 0.9754286 0.3184307 0.2410181
More information about the R-help
mailing list