MAIL: MAIL

In hwyneken/MAILPackage: Model-Averaged Inferential Learning

MAIL

Description

MAIL runs the Model-Averaged Inferential Learning method under different parameter settings

Usage

MAIL(
  XMat,
  yVec,
  splitOption,
  firstSOILWeightType,
  smallestModelWeightType,
  firstSOILPsi,
  smallestModelPsi,
  numSelectionIter = 10,
  sigma2EstFunc,
  trueSD = NULL,
  verbose = FALSE
)

Arguments

XMat	a n by p numeric matrix
yVec	a n by 1 numeric vector
splitOption	Mandatory - can take the values "Full" or "Split"
firstSOILWeightType	Mandatory - can take values "AIC", "BIC" or "ARM"
smallestModelWeightType	Mandatory - can take values "AIC", "BIC" or "ARM"
firstSOILPsi	Mandatory - can take any value in [0,1]
smallestModelPsi	Mandatory - can take any value in [0,1]
numSelectionIter	Optional - defaults to 10, must be an integer >= 1
sigma2EstFunc	Mandatory - this is a string of the function that will estimate the error variance using only XMat and yVec. We recommend using "LPM_AIC_CV_50Split". If the error variance is known, use "trueValue" here.
trueSD	Optional unless "trueValue" has given to the previous argument. This is where the user gives the assumed error standard deviation.
verbose	Optional: default is FALSE - set to TRUE if you want to see printed messages about MAIL's progress.

Details

The most important choice is whether or not use data splitting. The advantage of data splitting is to mitigate post selection changes to inference. The advantage of using all of the data is to reduce bias.

See Also

MAIL_Full and MAIL_Split for specific versions

LPM_AIC_CV_50Split for the recommended variance estimation method

hwyneken/MAILPackage documentation built on July 27, 2022, 12:46 a.m.