sigma_estim_condreg: Condition-Number Regularized Covariance Estimation

In antshi/CovEstim: (High-dimensional) Covariance Estimation

Description

Computes the condition-number regularized (CONDREG) estimator of the covariance matrix.

Usage

sigma_estim_condreg(
  data,
  k = NULL,
  k_seq = NULL,
  k_seq_len = 50,
  nfolds = 5,
  zeromean_log = FALSE
)

Arguments

data	an nxp data matrix.
k	a double, indicating the regularization parameter (or the maximum condition number for the estimated covariance matrix). Default value is NULL and the optimal k is found with a cross-validation (CV), optimizing the negative likelihood.
k_seq	a vector of doubles, specifying the grid of k values to search over within the CV. Default value is NULL and the sequence is generated in dependence of the sample covariance matrix, the user-supplied length and the minimum deviation ratio of the values.
k_seq_len	an integer, indicating the length of k_seq. Default value is 50.
nfolds	an integer, specifying the number of folds for the CV. Default value is 5.
zeromean_log	a logical, indicating whether the data matrix has zero means (TRUE) or not (FALSE). Default value is FALSE.

Details

The CONDREG estimator is elaborated in detail in \insertCitewon2013condition;textualCovEstim. More information on the functionality can be found in \insertCitecondregpackage;textualCovEstim.

Value

a list with the following entries

• a pxp estimated covariance matrix.

• an estimation specific tuning parameter, here the regularization parameter k.

References

\insertAllCited

Examples

data(sp200)
sp_rets <- sp200[,-1]
results_condreg <- sigma_estim_condreg(sp_rets)
sigma_condreg <- results_condreg[[1]]
param_condreg <- results_condreg[[2]]
sigma_condreg <- sigma_estim_condreg(sp_rets, k=100)[[1]]

antshi/CovEstim documentation built on Nov. 13, 2020, 2:25 p.m.