admix (version 2.1-3)

estimVarCov_empProcess: Variance-covariance matrix of the empirical process in an admixture model

Description

Estimate the variance-covariance matrix of some given empirical process, based on the Donsker correlation. Compute Donsker correlation between two time points (x,y) for some given empirical process with R code (another implementation in C++ is also available to speed up this computation).

Usage

estimVarCov_empProcess(
  x,
  y,
  obs.data,
  known.p = NULL,
  comp.dist = NULL,
  comp.param = NULL
)

Value

The estimated variance-covariance matrix.

Arguments

x

First time point considered for the computation of the correlation given the empirical process.

y

Second time point considered for the computation of the correlation given the same empirical process.

obs.data

Sample that permits to estimate the cumulative distribution function (cdf).

known.p

NULL by default (only useful to compute the exact Donsker correlation). The component weight dedicated to the unknown mixture component if available (in case of simulation studies)

comp.dist

NULL by default (only useful to compute the exact Donsker correlation). Otherwise, a list with two elements corresponding to component distributions (specified with R native names for these distributions) involved in the admixture model. All elements must be specified, for instance list(f='norm', g='norm').

comp.param

NULL by default (only useful to compute the exact Donsker correlation). Otherwise, a list with two elements corresponding to the parameters of the component distributions, each element being a list itself. The names used in this list must correspond to the native R argument names for these distributions. All elements must be specified, for instance list(f=NULL, g=list(mean=0,sd=1)).

Author

Xavier Milhaud

Examples

## Simulate data:
list.comp <- list(f1 = 'norm', g1 = 'norm')
list.param <- list(f1 = list(mean = 12, sd = 0.4),
                   g1 = list(mean = 16, sd = 0.7))
obs.data <- rsimmix(n=2500, unknownComp_weight=0.5, comp.dist=list.comp, comp.param= list.param)
## Compute the variance-covariance matrix of the corresponding empirical process:
t <- seq(from = min(obs.data$mixt.data), to = max(obs.data$mixt.data), length = 50)
S2 <- sapply(t, function(s1) {
                sapply(t, function(s2) {
                     estimVarCov_empProcess(x = s1, y = s2, obs.data = obs.data$mixt.data) })
                })
lattice::wireframe(S2)