Component-wise minimum variance semi-supervised regression.

<<<<<<< HEAD

AccuracyStdErrorEstimation(
  basis_labeled,
  basis_unlabeled,
  X_labeled,
  X_unlabeled,
  y,
  samp_prob,
  min_var_weight,
  beta_SL,
  beta_MV,
  beta_DR,
  resids_beta_SL,
  resids_beta_imp,
  resids_beta_dr,
  proj_dr,
  inverse_information,
  num_resamples = 500,
  threshold = 0.5
)

=======

Usage

AccuracyStdErrorEstimation(
  basis_labeled,
  basis_unlabeled,
  X_labeled,
  X_unlabeled,
  y,
  samp_prob,
  min_var_weight,
  beta_SL,
  beta_MV,
  beta_DR,
  resids_beta_SL,
  resids_beta_imp,
  resids_beta_dr,
  proj_dr,
  inverse_information,
  num_resamples = 500,
  threshold = 0.5
)

>>>>>>> e32a060aec520f495c0371cd7ee97e715754fa71

Arguments

basis_labeled: Basis matrix for labeled data set.
basis_unlabeled: Basis matrix for unlabeled data set.
X_labeled: Covariate matrix for labeled data set.
X_unlabeled: Covariate matrix for unlabeled data set.
y: Numeric outcome vector.
samp_prob: Numeric vector of weights.
min_var_weight: Minimum variance weight for semi-supervised estimate.
beta_SL: Supervised regression coefficient vector.
beta_MV: MinVar Semi-supervised regression coefficient vector.
beta_DR: Density ratio regression coefficient vector.
resids_beta_SL: Residuals from the supervised regression model.
resids_beta_imp: Residuals from the imputation model.
resids_beta_dr: Residuals from the density ratio estimator.
proj_dr: Projection from density ratio estimator.
inverse_information: Inverse information matrix.
num_resamples: Number of resamples.
threshold: Threshold for over misclassification rate.

<<<<<<< HEAD

Value

Pertrubed estimates.

=======

Value

Pertrubed estimates.

>>>>>>> e32a060aec520f495c0371cd7ee97e715754fa71