Data engineering and BI courses are free!
net(x, weights = c(v = 1, g = 1))matrix or data.frame with at least two rows and two
columns containing numeric data. The rows should represent a
sequence of sampling points with uniform intervals (e.g. a range of
years), but this is not checked. Each column is a time-series
spanning either the whole time range or a part of it.
numeric vector with two elements. Normally, variation
("v") and Gegenläufigkeit
("g") contribute to NET with equal weight. It is possible to
use different weights by setting them here. The names of the vector
are matched to c("v", "g") (see ‘Examples’). If no
names are given, the first element is the weight of variation.
list with the following components, in the same order as
described here:numeric vector containing
$NET[j]$. Row names of x (if any)
are copied here. numeric value $NET$, the
average of the "all" vector (NA values removed). This function computes the $NET$ parameter (Esper et
al., 2001). The overall $NET$ is an average of all
(non-NA) yearly values $NET[j]$, which are
computed as follows:
The yearly variation $v[j]$ is the standard deviation of the
measurements of a single year divided by their mean.
Gegenläufigkeit $1-G[j]$ is based
on one definition of Gleichläufigkeit
$G[j]$, similar to but not the same as what glk
computes. Particularly, in the formula used by this function (Esper
et al., 2001), simultaneous zero differences in two series are not
counted as a synchronous change.
The weights of $v[j]$ and $1-G[j]$ in the sum can
be adjusted with the argument weights (see above). As a
rather extreme example, it is possible to isolate variation or
Gegenläufigkeit by setting one of the weights
to zero (see ‘Examples’).
Esper, J., Neuwirth, B., and Treydte, K. (2001) A new parameter to evaluate temporal signal strength of tree-ring chronologies. Dendrochronologia, 19(1), 93–102.
data(ca533)
ca533.rwi <- detrend(rwl = ca533, method = "ModNegExp")
ca533.net <- net(ca533.rwi)
tail(ca533.net$all)
ca533.net$average
## Not run:
# ## Isolate the components of NET
# ca533.v <- net(ca533.rwi, weights=c(v=1,0))
# ca533.g <- net(ca533.rwi, weights=c(g=1,0))
# ## End(Not run)
Run the code above in your browser using DataLab