Bundled datasets

Both packages ship the example datasets used in the paper and vignettes. In R they are lazy-loaded data objects; in Stata they are loaded with stacked use.

R data(medicaid)   ↔︎   Stata stacked use medicaid

Usage

data(medicaid)      # then use the object `medicaid`
data(signflip)
stacked use medicaid, clear
stacked use signflip, clear

stacked use finds the bundled .dta from data/, the current directory, then the adopath. After a net install, fetch them once with net get stacked. The clear option replaces the data in memory.

Datasets

Name Rows Description R Stata
medicaid 714 State-year ACA Medicaid expansion panel; outcome uninsured. The paper’s headline example. yes yes
smoke 10,615 County-day smoke-plume / PM2.5 panel (2016); Date-keyed adoption. yes yes
acs_micro 69,820 0.3% ACS microdata sample; repeated cross-section, survey weight perwt, covariates. yes yes
signflip 72 Deterministic 12-unit teaching panel where naive TWFE flips the sign; Q-weights recover +2.4. yes yes
unparallel 120 Deterministic teaching panel with a differential pre-trend (parallel-trends violation). yes yes
pscore_stack — A pre-built stack with an estimated score phat, for the pscore demo. — Stata only

pscore_stack exists only in the Stata package (it cross-checks the pscore weights against an R-simulated golden stack); in R, build a stack and estimate the score inline as shown on the pscore page.

R ↔︎ Stata mapping

R Stata
data(medicaid) stacked use medicaid, clear
data(smoke) stacked use smoke, clear
data(acs_micro) stacked use acs_micro, clear
data(signflip) stacked use signflip, clear
data(unparallel) stacked use unparallel, clear
(build inline) stacked use pscore_stack, clear

Example

library(stacked)
data(medicaid)
str(medicaid)
Classes 'data.table' and 'data.frame':  714 obs. of  5 variables:
 $ state     : chr  "AL" "AL" "AL" "AL" ...
 $ statefip  : int  1 1 1 1 1 1 1 1 1 1 ...
 $ year      : int  2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 ...
 $ adopt_year: int  NA NA NA NA NA NA NA NA NA NA ...
 $ uninsured : num  0.196 0.214 0.23 0.225 0.216 ...
 - attr(*, ".internal.selfref")=<externalptr> 
head(medicaid)
   state statefip year adopt_year uninsured
1:    AL        1 2008         NA 0.1964095
2:    AL        1 2009         NA 0.2141095
3:    AL        1 2010         NA 0.2300015
4:    AL        1 2011         NA 0.2250678
5:    AL        1 2012         NA 0.2159348
6:    AL        1 2013         NA 0.2218695
stacked use medicaid, clear
describe
list in 1/6

See also