TSentryMinST: Two-Stage Minimum entry Test method for the loading vectors.
In feiyoung/SpagFainfer: Estimation and inference on high-dimensional group-sparse factor models

Description Usage Arguments Value Note Author(s) References See Also Examples

Conduct the simultaneous inference for a set of loading vectors as for the NUll hypothesises H01, where H01 assume the set of loading vectors are all zeroes.

1	TSentryMinST(X, S2, alpha=0.05, seed=1, sub.frac=0.5)

`X`	a `n`-by-`p` matrix, the observed data
`S2`	a matrix with two column whose rows represent bi-index, where the first index is between 1 and p, and the second index is between 1 and q, the set of H02.
`alpha`	a positive number, the significant level.
`seed`	a non-negative integer, the random seed.
`sub.frac`	a positive number between 0 and 1, the proportion of the sample used in stage I.

return a matrix with class "Min-test", row names 'chiq_test' , and column names 'CriticalValue', 'TestStatistic', 'reject_status', and 'p-value', including all the information about testing.

nothing

Liu Wei

Wei Liu, Huazhen Lin, Jin Liu (2020). Estimation and inference on high-dimensional sparse factor models.

factor, Factorm,TSrowMinST

  ### Example
  p <- 500; q <- 2
  dat <- gendata(n = 300, p = p, q=q)
  nzindMat <- indvec2matFun(which(dat$B0!=0), nrow=p)
  zindMat <- indvec2matFun(which(dat$B0==0), p)
  # ex1: H01 is false
  S2 <- rbind(c(1,2), c(1,3)); # all are nonzero loading entries
  TSentryMinST(dat$X, S2=S2, alpha=0.05, seed=10)
  # ex2: H02 is true: all are zero entries
  TSentryMinST(dat$X, S2=zindMat, alpha=0.05)
  # ex3: H02 is false: no zero entry.
  TSentryMinST(dat$X, S2=nzindMat, alpha=0.05)