Tests Doob's central thesis that imprisonment is decoupled from
crime by computing the Pearson correlation between the two time
series and (optionally) running Pettitt change-point on each.
Usage
decoupling_test(crime_series, imprisonment_series, years = NULL)
Arguments
- crime_series
Numeric vector of per-period crime rates.
- imprisonment_series
Numeric vector of per-period
imprisonment rates, same length as crime_series.
- years
Optional integer vector of period labels.
Value
A morie_result named-list.
Examples
set.seed(1)
crime <- rnorm(20)
imp <- 0.5 * crime + rnorm(20, sd = 0.5)
res <- decoupling_test(crime, imp, years = 2000:2019)
res$payload$r_pearson
#> [1] 0.6480564
