# Direction_fixedtuning: Constructs the projection direction with a fixed tuning... In SIHR: Statistical Inference in High Dimensional Regression

## Description

Constructs the projection direction, used for bias-correction, with a fixed tuning parameter

## Usage

 ```1 2 3 4 5 6 7 8``` ```Direction_fixedtuning( X, loading, mu = NULL, model = "linear", weight = NULL, deriv.vec = NULL ) ```

## Arguments

 `X` Design matrix, of dimension n x p `loading` Loading, of length p `mu` The dual tuning parameter used in the construction of the projection direction `model` The high dimensional regression model, either `linear` or `logistic` (default = `linear`) `weight` The weight vector of length n; to be supplied if `model="logistic"` (default=`NULL` when `model=linear`) `deriv.vec` The first derivative vector of the logit function at X\%*\%(`init.coef`), of length n ; to be supplied if `model="logistic"`. Here `init.coef` is the initial estimate of the regression vector. (default = `NULL` when `model=linear`)

## Value

 `proj` The projection direction, of length p

## Examples

 ```1 2 3 4 5 6``` ```n <- 100 p <- 400 X <- matrix(sample(-2:2,n*p,replace = TRUE),nrow = n,ncol = p) resol <- 1.5 step <- 3 Est <- Direction_fixedtuning(X,loading=c(1,rep(0,(p-1))),mu=sqrt(2.01*log(p)/n)*resol^{-(step-1)}) ```

SIHR documentation built on Oct. 7, 2021, 9:08 a.m.