>
>It seems they are using 10 minutes as the unit of measurement.
>If you wish to change it to hours you might want to use this instead:
>
> lh.hr <- ts(lh, start = 0, frequency = 6)
>
>so that
>
> cycle(lh.hr)
>
>starts out at 1.
Dear Gabor,
yes, it is very likely that the original dataset was with a 10 minutes
unit of measurement because plot(lh[1:44]) yields the same curve as
figure 1b in:
@article{DigglePJ1991,
title = {Nonparametric comparison of cumulative periodograms},
author = {Diggle, P.J. and Fisher, N.I.},
journal = {Applied Statistics},
volume = {40},
pages = {423-434},
year = {1991}
}
whose legend reads:
QUOTE
Fig. 1. Four time series of LH concentrations in blood samples: each
series consists of $n = 44$ values derived from blood samples taken at
10-min intervals from a healthy adult female (data collected by Dr A.
Murdoch): a) early follicular phase, first cycle; (b) early follicular
phase, second cycle; (c) late follicular phase, first cycle; (d) late
follicular phase, second cycle
UNQUOTE
The complete multiple time series with 48 values as in data(lh) was
deposited in statlib by P.J. Diggle himself:
http://lib.stat.cmu.edu/datasets/diggle
(note that the first one has 4 trailling 0 that look like NA which
may explain why there were only 44 values in the 1991 paper).
What I found confusing is that
plot(lh)
ends at 48 when the doc says that lh is a "regular time series".
That would be OK for me if lh was a simple vector, but lh is a
time series so I would expect it to have different tsp() parameters
to fit the doc.
Best,
Jean
--
Jean R. Lobry ([EMAIL PROTECTED])
Laboratoire BBE-CNRS-UMR-5558, Univ. C. Bernard - LYON I,
43 Bd 11/11/1918, F-69622 VILLEURBANNE CEDEX, FRANCE
allo : +33 472 43 27 56 fax : +33 472 43 13 88
http://pbil.univ-lyon1.fr/members/lobry/
______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
