EigenPrism: EigenPrism procedure for estimating and generating confidence...

In hychen-uic/TEV: Total Explained Variation

EigenPrism procedure for estimating and generating confidence intervals

Description

This function implements the EigenPrism procedure for estimating and generating confidence intervals for variance components in high-dimensional linear model.

Usage

EigenPrism(
  y,
  X,
  invsqrtSig = NULL,
  alpha = c(0.05),
  target = "beta2",
  zero.ind = c(),
  diagnostics = T
)

Arguments

y	response vector of length n
X	n by p design matrix. Columns are automatically centered and scaled to variance 1, and they cannot include one for the intercept terms.
invsqrtSig	if columns of X are not independent, p by p positive definite matrix which is the square-root of the inverse of Sig, where Sig is the correlation matrix of the X. Default is identity
alpha	significance level for confidence interval. Default is 0.05
target	target of estimation/inference: options include "beta2", "sigma2", or "heritability"
zero.ind	vector of which indices of the weight vector w to constrain to zero. Default is none
diagnostics	Boolean variable indicating whether to generate the diagnostic plots for V_i, lambda_i, and w_i. Default is TRUE.

Details

This function is a copy of Jansen's R program. It implements the EigenPrism procedure for estimating and generating confidence intervals for variance components in high-dimensional linear model.

Value

Estimate of the proportion of explained variation and 100*(1-alpha)\

References

Janson, L., Barber, R. F., Candes, E. (2017). EigenPrism: inference for high-dimensional signal-to-noise ratios. Journal of Royal Statistical Society, Ser. B., 79, 1037-1065.

Lucas Janson. http://lucasjanson.fas.harvard.edu/code/EigenPrism.R.

Examples

## Not run: EigenPrism(y,x)

hychen-uic/TEV documentation built on Jan. 24, 2025, 9:14 p.m.