Description Usage Arguments Details Value Author(s) References See Also Examples
diagnostic_dawson
take two vectors
(assumed to be time series) and
calculates the following objective functions to compare them:
correalation, Nash Sutcliffe efficiency, ratio of the integral,
lagtime (maximum of the cross correlation), the number of timesteps
with opposite sign of the derivative, the highest ratio between
recession coefficients and the root mean square error, as well as
the ones listed in Dawson 2007.
diagnostic_window
calcualtes these measures for a
part of the time series only. It is used internally by
diagnostic_series
takes this a step further by
calculating the above measures for a gliding window along the time
series and calculating additional measures.
Similar to diagnostic
, the function
takes two vectors (assumed to be time series) and
calculates a number of objectives compare them.
In contrast to the more simple diagnostic
, the same objectives
are applied to a gliding window and a few additional objectives
are calcualated: the ratio of the derivatives, the ratio of the
recession coefficients for each time step and the current quantile
of the residuals.
1 2 3 4 5 6 7 8 | diagnostic_window(position, window.size, measured,
modelled, use_qualV = FALSE,
diff.ecdf=NA)
diagnostic_series(measured, modelled, window.size,
step.size = 1, integral_correction = FALSE,
use_qualV = FALSE)
diagnostic_dawson(modelled, measured, p=NA, m=NA, additional=TRUE,
use_qualV=FALSE, diff.ecdf=NA )
|
modelled |
Modelled time series or array with dimension |
measured |
Measured time series |
position |
Index from where to start the calculation |
window.size |
Number of time steps to include |
step.size |
Size of the steps defining the number of scores to be calculating along the time series. For example, with a value of 5 every fifth value is included |
integral_correction |
Boolean. If true, the ratio of the integrals is divided by the total ratio of the entire integral. This way, relative integral errors can be detected. |
p |
The number of free parameters in each model - required to calculate AIC and BIC |
m |
The number of data points that were used in the model calibration - required to calculate AIC and BIC |
additional |
Boolean, indicating whether to calculate additional measures to the ones defined in Dawson 2007 |
use_qualV |
Boolean, indicating whether to calculate the additional measures defined in Jachner 2007 |
diff.ecdf |
|
For more details on the objectives, see the see-also-section
A data frame with the described objectives
Dominik Reusser
Dawson, C. W.; Abrahart, R. J. & See, L. M. HydroTest: A web-based toolbox of evaluation metrics for the standardised assessment of hydrological forecasts Environmental Modelling & Software, 2007 , 22 , 1034-1052
Jachner, S.; van den Boogaart, K. G. & Petzoldt, T. Statistical Methods for the Qualitative Assessment of Dynamic Models with Time Delay (R Package qualV) Journal of Statistical Software, 2007 , 22 , 1-30
qualV
cor
,
nashS
,lagtime
,count.diff.direction.error
,k_rel
1 2 3 4 5 6 7 8 9 10 11 12 13 14 | data(example.peaks,package="tiger")
plot(reference.peak, type="l")
lines(example.peaks[1,], lty=2)
diagnostic_dawson(measured = reference.peak, modelled = example.peaks[1,])
#first half only
diagnostic_window(measured=reference.peak, modelled=example.peaks[1,],
position = 1, window.size = 45 )
#gliding window for 20 time steps
diagnostic_series(measured=reference.peak, modelled=example.peaks[1,],
window.size = 20 )
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.