RVsd | R Documentation |
This function computes: (1). the proportion of the explained variation (2). The explained variation adjusting for the correlation in covariates with possible suplementary data
RVsd(
y,
x,
xsup = NULL,
lam = 1,
niter = 1,
alpha = c(0.05),
know = "no",
KV = array(0, c(3, 100)),
nrep = 1000
)
y |
outcome: a vector of length n. |
x |
covariates: a matrix of nxp dimension. |
xsup |
supplementary covariates: a matrix of Nxp dimension. when xsup=NULL and n>p, it performs least-square with the simulation variance estimate |
lam |
parameter for altering the weighting matrix. |
niter |
number of iterations for updating lam. |
alpha |
a vector of type I errors used to generate (1-alpha)confidence intervals. |
know |
if KV is known, options include "yes" and "no". Default is "no". |
KV |
a matrix with 3 rows and 100 collums corresponding to different lambda values. the first component of the row vector KV=kappa_1, the second=kappa_2, the third=kappa_3 |
nrep |
Monte Carlo sample size for computinging KV. |
The estimation approach does not assume independent covariates and can deal
with the case n\le p
. But require the sample sizes of x and X combined be greater than p.
Estimate of the proportion of explained variation, variance estimates under normality and non-normality assumptions, and confidence intervals under normality and non-normality assumptions.
Chen, H. Y., Li, H., Argos, M., Persky, V. W., and Turyk, M. (2022). Statistical Methods for Assessing Explained Variation of a Health Outcome by Mixture of Exposures. International Journal of Environmental Research and Public Health.
Chen H. Y, Zhang, B. and Pan, G.(2023).Estimation and inference on explained variation with possible supplementary data, Manuscript.
## Not run: RVsd(y,x,xsup,lam=0.1,niter=3,alpha=c(0.1,0.05,0.01),KV=rep(0,3),know="no",nrep=1000)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.