https://github.com/cran/nacopula
Tip revision: 161411bb86f97e5a8bd89091cd61d03a33c2761a authored by Martin Maechler on 06 February 2012, 00:00:00 UTC
version 0.8-0
version 0.8-0
Tip revision: 161411b
TODO
#### Major (to fix before next release) ########################################
(*) bug fixes for estimation methods!
#### Intermediate #####################################################################
(MM) R/special-func.R & man/Bernoulli.Rd: Research for better algorithms <--> Rmpfr
(MM) please take a look at gtrafouni() in ./R/gof.R: Can "Remillard" (and "Genest")
be made faster? Especially "Remillard" is slow (although B Remillard told me that his
method is faster than C. Genest's.)
#### Minor #####################################################################
(*) write nice little function that provides confidence intervals for tau and lambda's
(*) fix:
cop <- onacopulaL("Frank", list(200, 1:2)); rnacopula(100, cop)
maybe use limit copulas also for other families (upper frechet bound)
(*) add an ASCII-plot for the hierarchical structure;
Actually, define setAs("nacopula", "dendrogram", ....) and then use
dendrogram print()ing.
(*) design a function that gets an outer nacopula and returns a matrix of
numbers which correspond to the different sectors; this can be used, e.g.,
for building a [default] color-matrix for splom2 or for estimation purposes etc.
(*) it would be good to have tryCatch.W.E in the package [call it tryCatchWE --
easier to type]
Maybe one can also directly return a list with "value", "error", and "warning"?
[So far, "value" can also contain the errors => a bit unhandy, but maybe the
only "elegant" general solution]
(*) From time to time (well, whenever Marius has.. ;-)) check the .Rd files
for overlong lines, i.e. apply
for f in *Rd; do printf '%20s: ' $f ; printf '%4d\n' `maxcol $f` ; done
#### Long-term maybe ###########################################################
(*) implement Khoudraji or Liebscher trafo incl. density etc.
(*) allow families to have outer power parameters (outer power families)
(*) be compatible with package copula
(*) include pairwise estimators for nACs
(*) FIXME in enacopula (in estimation.R): return an object?
(*) can we create a function for the GIG generator similar to opower()?