# [R] intervals from cut() as numerics?

From: Gavin Simpson <gavin.simpson_at_ucl.ac.uk>
Date: Sat 20 May 2006 - 19:01:04 EST

Hi,

Given some example data:

```dat <- seq(4, 7, by = 0.05)
x <- sample(dat, 30)
y <- sample(dat, 30)
```

error <- x - y

I have broken the rage of x into 10 groups and I can calculate the bias (mean(error)) for each of these 10 groups:

groups <- cut(x, breaks = 10)
max.bias <- aggregate(error, list(group = groups), mean) max.bias

group x

```1    (4,4.3] -0.7750000
2  (4.3,4.6] -0.9750000
3  (4.6,4.9] -0.6750000
4  (4.9,5.2] -0.9125000
5  (5.2,5.5]  1.0500000
6  (5.5,5.8]  0.3166667
7  (5.8,6.1]  0.1375000
8  (6.1,6.4]  1.1833333
9  (6.4,6.7]  0.2500000
```

10 (6.7,7] 1.7000000

Which is fine. Now I am producing a plot of the residuals vs observed and I want to draw line segments from e.g. (4,4.3] at height = -0.7750000, but for all 10 groups, like so:

plot(x, error, type = "n")
abline(h = 0, col = "grey")
panel.smooth(x, error)
arrows(4, -0.775, 4.3, -0.775, length = 0.05, angle = 90, code = 3)

The problem is getting the range/interval for each group from (4,4.3], so I can automate this.

Anyone think of a way to do this - happy to change the way the groups are generated if cut() is not the right tool for the job here.

Many thanks,

Gav

```--

%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
*Note new Address and Fax and Telephone numbers from 10th April 2006*

%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
Gavin Simpson                     [t] +44 (0)20 7679 0522
ECRC                              [f] +44 (0)20 7679 0565
UCL Department of Geography
Pearson Building                  [e] gavin.simpsonATNOSPAMucl.ac.uk
Gower Street
London, UK                        [w] http://www.ucl.ac.uk/~ucfagls/cv/
WC1E 6BT                          [w] http://www.ucl.ac.uk/~ucfagls/

%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%

______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help