ps.summary {twang}R Documentation

Computes balance measures

Description

Computes balance measures (mean differences and KS statistics) for a particular covariate and a set of propensity score weights. This function is not intended to be called directly by the user but is used by other functions in the package.

Usage

ps.summary(x, t, w, get.means = TRUE, get.ks = TRUE, 
           na.action = c("level", "exclude", "lowest")[1], 
           collapse.by.var = FALSE)
ps.summary.f(x, t, w, get.means = TRUE, get.ks = TRUE, 
             na.action = c("level", "exclude", "lowest")[1], 
             collapse.by.var = TRUE)
ps.summary.n(x, t, w, get.means = TRUE, get.ks = TRUE, 
             na.action = c("level", "exclude", "lowest")[1], 
             collapse.by.var = FALSE)

Arguments

x a vector containing the data for a single covariate
t a vector of the same length as x with the 0/1 treatment assigniments
w a vector of the same length as x with the weights
get.means if TRUE, mean comparisons are computed
get.ks if TRUE, the KS statistics are computed
na.action a string indicating the method for handling missing data
collapse.by.var if TRUE, then statistics computed for factors are collapsed across the levels

Details

ps.summary dispatches ps.summary.n or ps.summary.f depending on whether x is a numeric vector or a factor.

Value

Returns a data frame containing the balance information.

tx.mn The mean of the treatment group
tx.sd The standard deviation of the treatment group
ct.mn The mean of the control group
ct.sd The standard deviation of the control group
std.eff.sz The standardized effect size, (tx.mn-ct.mn)/tx.sd
stat the t-statistic for numeric variables and the chi-square statistic for continuous variables
p the p-value for the test associated with stat
ks the KS statistic
ks.pval the KS p-value computed using the analytic approximation, which does not necessarily work well with a lot of ties


get.means and get.ks manipulate the inclusion of certain columns in the returned result.

See Also

bal.stat, ks.stat, es.stat

Examples

treat <- rbinom(100,1,0.5)
w     <- rexp(100)

# categorical data
x.cat <- factor(sample(letters[1:3],size=100,replace=TRUE))
ps.summary.f(x.cat,treat,w)

# numeric data
x.num <- rnorm(100)
ps.summary.n(x.num,treat,w)

# or let ps.summary figure out which to call
ps.summary(x.num,treat,w)

[Package twang version 1.0-1 Index]