beta.r.compute: Computes the restricted estimator in the 'fsst' procedure
In conroylau/lpinfer: An R Package for Inference in Linear Programs

beta.r.compute

R Documentation

Computes the restricted estimator in the `fsst` procedure

This function computes the restricted estimator \widehat{\bm{β}}^r_n in the fsst procedure.

beta.r.compute(
  n,
  lpmodel,
  beta.obs.hat,
  beta.tgt,
  beta.n,
  beta.star,
  omega.i,
  indicator,
  solver
)

`n`	The sample size. This is only required if `data` is omitted in the input.
`lpmodel`	The `lpmodel` object.
`beta.obs.hat`	The sample estimator \widehat{\bm{β}}_{\mathrm{obs}, n} based on the given information in `lpmodel` (and `data` if applicable).
`beta.tgt`	The value to be tested.
`beta.n`	The sample \widehat{\bm{β}}_n vector that is defined as \widehat{\bm{β}}_n \equiv (\widehat{\bm{β}}_{{\rm obs},n}, \bm{β}_{{\rm shp},n}, \bm{β}_{{\rm tgt}})'.
`beta.star`	The starred version of the `beta.n` vector, i.e. \widehat{\bm{β}}^\star_n.
`omega.i`	The matrix \widehat{\bm{Ω}}^i_n, i.e. the regularized matrix for \widehat{\bm{Σ}}^{β^\star}_{n,\bar{ρ}}.
`indicator`	A binary variable that equals to 1 for d ≥q p and equals to 0 for d < p.
`solver`	The name of the linear and quadratic programming solver that is used to obtain the solution to linear and quadratic programs. The solvers supported by this package are `cplexAPI`, `gurobi`, `limSolve` and `Rcplex`.