boot_default

a function for weights calculation (nn / non_parm).

func

outcome after exponential transformation (if needed).

Y_pos

X_std

numeric data frame or matrix of factors to control for. these are factors
that we can't consider while looking for the optimal intervention
(e.g. race).

control

number of quantiles to use when calculating CDF distance.

n.quant

the lagrange multiplier. also known as the shadow price of an
intervention.

lambda

distance penalty for the nearest-neighbors method.

sigma

for the nearest-neighbors method; if the number of examples in each
control group is smaller than grp.size, performs weight adjustment
using <code><a rd-options="optinterv" href="/link/wgt_adjust?package=optinterv&version=0.1.0&to=optinterv" data-mini-rdoc="optinterv::wgt_adjust">wgt_adjust</a></code>. else,
calculate weights seperatly for each control group.

grp.size

number of bootstrap replications to use for the standard errors /
confidence intervals calculation.

n.boot

logical. if TRUE, returns only \(E(X | I=1) - E(X | I=0)\) as an estimate.
this estimate is used by <code><a rd-options="" href="/link/optint_by_group?package=optinterv&version=0.1.0" data-mini-rdoc="optinterv::optint_by_group">optint_by_group</a></code>.

quick

Bootstrap (default)
Bootstrap function for the non-parametric and the nearest neighbor methods


Provides both parametric and non-parametric estimates of the
correlates of some desired
outcome (e.g. test scores, income) using a new method proposed by
Danieli, Devi and Fryer (2019). This method relaxes the assumption
that one can alter individual characteristics in any way the data
suggest is optimal, and so can be used anytime one wants to use observational
data to better optimize social experiments designed to increase some desired
outcome.

Elad Guttman

optinterv

Optimal Intervention

Oren Danieli

boot_default function

for the nearest-neighbors method; if the number of examples in each
control group is smaller than grp.size, performs weight adjustment
using <code><a rd-options='optinterv' href='wgt_adjust'>wgt_adjust</a></code>. else,
calculate weights seperatly for each control group.

logical. if TRUE, returns only \(E(X | I=1) - E(X | I=0)\) as an estimate.
this estimate is used by <code><a rd-options='' href='optint_by_group'>optint_by_group</a></code>.

boot_default: Bootstrap (default) Bootstrap function for the non-parametric and the nearest neighbor methods

Description

Usage

Arguments

Value