R/ittReg.R

Defines functions ittReg

Documented in ittReg

#' Intent to treat regression
#' @param y outcome, default is outcome
#' @param tx treatment assignment variable
#' @param x covariates, default named covariates
#' @param data data frame, default named df
#' @author William Murrah
#' @export
ittReg <- function(y=outcome,tx=tx.assign, x=covariates,
                    data=df, ...){
  # Creates an intent-to treat model with the lm function. The outcome is
  # assumed to be continuous and is scaled in the analysis.
  #
  # Args:
  #   y:         Character vector of length = 1 that is the column name of the
  #              outcome variable.
  #   tx.assign: Character vector of length = 1 that is the column name of the
  #              treatment assignment variable (randomized variable).
  #   x:         Character vector of any length that contains the column names
  #              of the covariates.
  #   data:      The data frame containing the variables referenced by the
  #              previous arguments.
  # Returns:
  #   An object of class 'lm'.
  # Create formula to be passed to 'lm'.
  fml <- as.formula(paste('scale(',y,')', ' ~ ', tx, ' + ',
                          paste(x, collapse=' + ')))

  # Create lm model object.
  mod <- lm(fml, data=df)
  return(mod)
}
wmmurrah/cpar documentation built on May 4, 2019, 9:45 a.m.