SearchingSampling.sumstats: Searhing-and-Sampling method

In MinhaoYaooo/MR.TSHT: Two Sample Mendelian Randomization using Two-stage Hard Thresholding with Voting

Searhing-and-Sampling method

Description

Get a possible confidence interval using searching method

Usage

SearchingSampling.sumstats(
  ITT_D,
  ITT_Y,
  SE_D,
  SE_Y,
  n1,
  n2,
  CI.init = NULL,
  a = 0.6,
  Sampling = TRUE,
  rho = NULL,
  M = 1000,
  prop = 0.1
)

Arguments

ITT_D	a numeric vector of GWAS summary statistics of the treatment
ITT_Y	a numeric vector of GWAS summary statistics of the outcome
SE_D	a numeric vector of standard errors of ITT_D
SE_Y	a numeric vector of standard errors of ITT_Y
n1	the sample size of GWAS summary statistics of the treatment
n2	the sample size of GWAS summary statistics of the outcome
CI.init	initial interval for beta. If NULL, it will be generated automatically. (default=NULL)
a	grid size for constructing beta grids (default=0.6)
Sampling	if TRUE, use the proposed sampling method; else use the proposed searching method. (default=TRUE)
rho	a numeric scalar denoting thresholding level used in the sampling property
M	sampling times. (default=1000)
prop	proportion of intervals kept when sampling. (default=0.1)

Value

CI	a numeric matrix denoting the confidence intervals for betaHat constructed by valid candidates for betaHat.
rule	a boolean scalar denoting whether the identification condition is satisfied or not.
VHat	valid instruments
CI.init	initial confidence interval for searching and sampling
TSHT.out	standard MR-TSHT output

MinhaoYaooo/MR.TSHT documentation built on May 22, 2022, 7:19 a.m.