generate_blocked_data_linear: Make data from a linear model specification

In lmiratrix/blkvar: ATE and Treatment Variation Estimation for Blocked and Multisite RCTs

Make data from a linear model specification

Description

Function that, given a covariate vector X, returns a dataframe of potential outcomes and blocks.

Generate data, form_blocks_from_continuous, and randomize within block and generate observed potential outcomes

Usage

generate_blocked_data_linear(
  X = c(0, 2, 3, 19, 20, 21, 24, 31, 32, 40, 41, 43, 45, 55, 60, 65),
  a = 0,
  b = 0,
  ATE = 0.2,
  d = 0
)

generate_blocked_data_obs_linear(
  X = c(0, 2, 3, 19, 20, 21, 24, 31, 32, 40, 41, 43, 45, 55, 60, 65),
  p = 0.5,
  a = 0,
  b = 0,
  ATE = 0.2,
  d = 0,
  method = c("small", "pair", "big", "none")
)

Arguments

X	vector of indiviual level covariates
a	Intercept of Y0
b	Main effect of X
ATE	Average Tx
d	Interaction effect term.
p	Proportion of units treated (as close as possible given block sizes)
method	How to block

Details

This uses the model: Y = a + bX + ATE Z + d X ATE + epsilon (It will standardize X for this model, and then standardize Y0, Y1 so sd(Y0) = 1.)

Value

Dataframe with original potential outcomes and observed outcome based on random assigment.

lmiratrix/blkvar documentation built on Nov. 18, 2024, 1:27 p.m.