simnul

Simulates the test statistic, under independence

Estimation of copula using ranks and subsampling. The main feature of this method is that simulation studies show a low sensitivity to dimension, on realistic cases.

Jerome Collet

subrank

Computes Copula using Ranks and Subsampling

simnul function

<dl> <dt>sampsize</dt>
<dd>sample size</dd> <dt>dimension</dt>
<dd>sample dimension</dd> <dt>subsampsizes</dt>
<dd>vector of sub-sample sizes</dd> <dt>sampnum</dt>
<dd>number of samples</dd> <dt>KL</dt>
<dd>if TRUE, returns the Kullback-Leibler divergence with the independent case, if FALSE, the L2 distance. There is no re-normalization, contrary to what happens for <code>simany</code>.</dd> <dt>nbsafe</dt>
<dd>the ratio between the number of sub-samples and the cardinality of the discretized copula.</dd> <dt>nthreads</dt>
<dd>number of number of threads, assumed to be strictly positive. For "full throttle" computations, consider using parallel::detectCores()</dd></dl>

Arguments

Author

Test statistic distribution under independence hypothesis — simnul

<dl>

 <dt>sampsize</dt>
<dd>sample size</dd>

 <dt>dimension</dt>
<dd>sample dimension</dd>

 <dt>subsampsizes</dt>
<dd>vector of sub-sample sizes</dd>

 <dt>sampnum</dt>
<dd>number of samples</dd>

 <dt>KL</dt>
<dd>if TRUE, returns the Kullback-Leibler divergence with the independent case, if FALSE, the L2 distance. There is no re-normalization, contrary to what happens for <code>simany</code>.</dd>

 <dt>nbsafe</dt>
<dd>the ratio between the number of sub-samples and the cardinality of the discretized copula.</dd>

 <dt>nthreads</dt>
<dd>number of number of threads, assumed to be strictly positive. For "full throttle" computations, consider using parallel::detectCores()</dd>

</dl>

simnul: Test statistic distribution under independence hypothesis

Description

Usage

Value

Arguments

Author

Examples