wdd_harm: Combines Multiple WDD Tests

In ptools: Tools for Poisson Data

Combines Multiple WDD Tests

Description

Combines multiple weighted displacement difference tests into one final weighted harm metric.

Usage

wdd_harm(est, se, weight, alpha = 0.1, silent = FALSE)

Arguments

est	vector with WDD estimates (e.g. difference in crime counts for treated vs controls)
se	vector with standard errors for WDD estimates
weight	vector with weights to aggregate results
alpha	scaler alpha level for confidence interval (default 0.1)
silent	boolean, do not print stat messages (default FALSE)

Details

This test combines multiple wdd estimates with different weights. Created to combine tests for crime harm weights.

Value

A length 5 vector with names:

• HarmEst, the combined harm estimate

• SE_HarmEst its standard error

• Z, the Z-score

• and the lower and upper confidence intervals, LowCI and HighCI, for whatever alpha level you specified.

See Also

wdd() for estimating the individual wdd outcomes

Examples

# Creating wdd tests for three different crimes and combining
rob <- wdd(c(20,20),c(20,10))
burg <- wdd(c(30,30),c(25,20))
theft <- wdd(c(80,60),c(70,20))
dat = data.frame(rbind(rob,burg,theft))
# passing those columns now to the wdd_harm function
harm_weights <- c(10,5,1)
wdd_harm(dat$Est_Local,dat$SE_Local,harm_weights)

ptools documentation built on Feb. 16, 2023, 10:40 p.m.