calibrate: Estimate Observations of X from a Given 'Y'

In ryan-heslin/RegLesson: Linear Regression from Scratch

Estimate Observations of X from a Given Y

Description

Estimate Observations of X from a Given Y

Usage

calibrate(
  Y,
  Y_0,
  betas,
  Y_bar = mean(Y),
  X,
  m = 1,
  n = NROW(Y),
  p = NROW(betas),
  MSE,
  alpha = 0.05,
  simultaneous = FALSE,
  intercept_fitted = TRUE
)

Arguments

Y	Vector of Y observations used to fit the model
Y_0	Hypothetical Y observations on which to predict X
betas	Estimated regression coefficients.
Y_bar	Mean of the response vector Y. Will be computed if not provided.
X	Design matrix used to fit a model.
m	Number of new observations at each level of Y_0 to predict.
n	Number of rows in the design matrix.
p	Number of parameters in the model.
MSE	Estimated mean squared error.
alpha	Alpha level for confidence intervals.
simultaneous	Logical. Should Bonferroni-corrected simultaneous intervals be constructed for each prediction? Default FALSE.
intercept_fitted	Logical. Does the model contain an intercept parameter? Default FALSE.

Details

this function does inverse prediction, finding observations of X given observations of Y.

Value

3 \times h matrix of predictions with interval bounds, where h is n the number of observations predicted.

ryan-heslin/RegLesson documentation built on Aug. 5, 2022, 9:03 p.m.