R/diagnostic.R, R/diagnostic.fitmodel.R
diagnostic.RdS3 method for evaluating diagnostics of a fitted LRE object
Produces a series of standard suite of diagnostics for examining the fit of a loads model.
diagnostic(object)
# S3 method for class 'fitmodel'
diagnostic(object)object of class "fitmodel" resulting from
the FitModel function.
Produces the following diagnostics plots:
residuals versus predicted points should appear randomly
distributed along the line at zero.
normal Quantile-Quantile (Q-Q) plot compares the quantiles
from the theoretical distribution (i.e. normal) to that observed.
If the 2 distributions are similar, then the points in the Q-Q plot
will approximately lie on the line. Som deviation at the tails is
acceptable.
histogram of residuals should appear roughly symmetric and bell
shaped. No skewness (elongation of the tails) or kurtosis (thickening
of the tails) should be present.
predicted versus observed points should appear close to the
diagonal 0-1 line.
predict versus obsered (WY) plots of predicted versus observed
for each water year.
residuals versus predicted (WY) plots of of residuals versus
predicted for each water year.
@importFrom "stats" "resid" @importFrom "stats" "lag" @importFrom "mgcv" "predict.gam" @importFrom "nlme" "predict"
@references Kuhnert, P.M., Henderson, B.L., Lewis, S.E., Bainbridge, Z.T., Wilkinson, S.N. and Brodie, J.E. (2012) Quantifying total suspended sediment export from the Burdekin River catchment using the loads regression estimator tool, Water Resources Research, 48, W04533,doi:10.1029/2011WR011080.
@seealso FitModel, qqnorm,, qqline
@keywords loads, diagnostics
Series of diagnostic plots for examining model fit.