startcost: Initialize a Distance Matrix.

In tightenBlock: Tightens an Observational Block Design by Balanced Subset Matching

Initialize a Distance Matrix.

Description

Creates an distance matrix of zeros of dimensions compatible with the treatment indicator vector z.

Usage

startcost(z)

Arguments

z

A vector with z[i]=1 if individual i is treated or z[i]=0 if individual i is control. The rows of costmatrix refer to treated individuals and the columns refer to controls. Although not strictly required, it is best that z has names that are the same as the names of the data frame dat that will be used in matching.

Value

A matrix of zeros with sum(z) rows and sum(1-z) columns. If z has names, then they become the row and column names of this matrix.

Author(s)

Paul R. Rosenbaum

Examples

# Although names are not required, they aid clarity.  The cost matrix
# has sum(z) rows and sum(1-z) columns, where z is a binary vector.
# If needed, it is easier to connect an entry in z to a row or column in
# cost if they have the same name.  If z has names, the data matrix should
# have the same names.
data(aHDLt)
rownames(aHDLt)<-aHDLt$SEQN
z<-aHDLt$z
names(z)<-aHDLt$SEQN
dist<-startcost(z)
dist[1:3,1:9]

tightenBlock documentation built on May 29, 2024, 1:16 a.m.