cvtunerange: Tuning parameter range

In lasandrall/SELPCCA: Sparse Canonical Correlation Analysis for Associating Mutiple High Dimensional Data

Description

Obtain upper and lower bounds of tuning parameters for each canonical correlation vector. It is recommended to use cvselpscca to choose optimal tuning paramters for each dataset.

Usage

cvtunerange(Xdata1=Xdata1,Xdata2=Xdata2,ncancorr=ncancorr,
            CovStructure="Iden",standardize=TRUE)

Arguments

Xdata1	A matrix of size n \times p for first dataset. Rows are samples and columns are variables.
Xdata2	A matrix of size n \times q for second dataset. Rows are samples and columns are variables.
ncancorr	Number of canonical correlation vectors. Default is one.
CovStructure	Covariance structure to use in estimating sparse canonical correlation vectors. Either "Iden" or "Ridge". Iden assumes the covariance matrix for each dataset is identity. Ridge uses the sample covariance for each dataset. See reference article for more details.
standardize	TRUE or FALSE. If TRUE, data will be normalized to have mean zero and variance one for each variable. Default is TRUE.

Details

The function will return tuning ranges for sparse estimation of canonical correlation vectors. To see the results, use the “$" operator.

Value

TauX1range	A ncancorr \times 2 matrix of upper and lower bounds of tuning parameters for each canonical correlation vector for first dataset.
TauX2range	A ncancorr \times 2 matrix Upper and lower bounds of tuning parameters for each canonical correlation vector for second dataset.

References

Sandra E. Safo, Jeongyoun Ahn, Yongho Jeon, and Sungkyu Jung (2018) , Sparse Generalized Eigenvalue Problem with Application to Canonical Correlation Analysis for Integrative Analysis of Methylation and Gene Expression Data. Biometrics

See Also

cvselpscca, multiplescca

Examples

#see example in multiplescca

lasandrall/SELPCCA documentation built on June 8, 2020, 12:38 a.m.