create_list_from_scratch_overall

This function takes in a n-by-p matrix of observed covariates,
a length-n vector of treatment indicator, a caliper, and construct
a possibly sparse list representation of the distance matrix with
Mahalanobis distance. Note that this function is of limited interest
to most users.

Multivariate matching in observational studies typically has two goals: 1. to construct
treated and control groups that have similar distribution of observed covariates and 2. to produce
matched pairs or sets that are homogeneous in a few priority variables. This packages implements a
network-flow-based method built around a tripartite graph that can simultaneously achieve both goals.
The package also implements a template matching algorithm using a variant of the tripartite graph
design. A brief description of the workflow and some examples are given in the vignette. A more elaborated
tutorial can be found at <https://www.researchgate.net/publication/359513837_Tutorial_for_R_Package_match2C>.

Bo Zhang

match2C

Match One Sample using Two Criteria

create_list_from_scratch_overall function

<dl><dt>Z</dt>
<dd>A length-n vector of treatment indicator.</dd>
<dt>X</dt>
<dd>A n-by-p matrix of covariates.</dd>
<dt>exact</dt>
<dd>A vector of strings indicating which variables are to be exactly matched.</dd>
<dt>soft_exact</dt>
<dd>If set to TRUE, the exact constraint is enforced up to a large penalty.</dd>
<dt>p</dt>
<dd>A length-n vector on which a caliper applies, e.g. a vector of propensity score.</dd>
<dt>caliper_low</dt>
<dd>Size of caliper_inf.</dd>
<dt>caliper_high</dt>
<dd>Size of caliper_sup.</dd>
<dt>k</dt>
<dd>Connect each treated to the nearest k controls</dd>
<dt>penalty</dt>
<dd>Penalty for violating the caliper. Set to Inf by default.</dd>
<dt>dist_func</dt>
<dd>A function used to calculate distance</dd></dl>

Arguments

Create a sparse list representation of treated-to-control distance
matrix with a fixed number caliper with L1-distance. — create_list_from_scratch_overall

<dl>

<dt>Z</dt>
<dd>A length-n vector of treatment indicator.</dd>


<dt>X</dt>
<dd>A n-by-p matrix of covariates.</dd>


<dt>exact</dt>
<dd>A vector of strings indicating which variables are to be exactly matched.</dd>


<dt>soft_exact</dt>
<dd>If set to TRUE, the exact constraint is enforced up to a large penalty.</dd>


<dt>p</dt>
<dd>A length-n vector on which a caliper applies, e.g. a vector of propensity score.</dd>


<dt>caliper_low</dt>
<dd>Size of caliper_inf.</dd>


<dt>caliper_high</dt>
<dd>Size of caliper_sup.</dd>


<dt>k</dt>
<dd>Connect each treated to the nearest k controls</dd>


<dt>penalty</dt>
<dd>Penalty for violating the caliper. Set to Inf by default.</dd>


<dt>dist_func</dt>
<dd>A function used to calculate distance</dd>

</dl>

create_list_from_scratch_overall: Create a sparse list representation of treated-to-control distance matrix with a fixed number caliper with L1-distance.

Description

Usage

Value

Arguments