eval_metrics_simu_group: Evluating pESCA model when simulated parameters are avaliable

In YipengUva/RpESCA: R implementaion of pESCA models with various concave penalties

Description

This function will evaluate the the performance of the constructed pESCA model with group penalty when the simulated parameters are avaliable.

Usage

eval_metrics_simu_group(mu, A, B, S, ds, simulatedData)

Arguments

mu	estimated offset term in column vector form
A	estimated score matrix
B	estimated loading matrix
S	estimated group sparse pattern on B
ds	a vector contains the number of variables in multiple data sets
simulatedData	the output of function dataSimu_group_sparse

Value

This function returns a list contains

• RVs_structs: a vector contains the RV coefficients in estimating the global common (C123), local common (C12, C13, C23) and distinct structures (D1, D2, D3);

• Ranks_structs: a vector contains the ranks of estimated C123, C12, C13, C23, D1, D2, D3;

• RMSEs_params: the relative mean squared errors (RMSEs) in estimating the simulated parameters Θ, Θ_1, Θ_2, Θ_3, μ

Examples

## Not run: eval_metrics_simu_group(mu,A,B,S,ds,simulatedData)

YipengUva/RpESCA documentation built on July 2, 2019, 6:41 p.m.