Skip to content

Instantly share code, notes, and snippets.

@jwbowers
Created April 16, 2020 20:09
Show Gist options
  • Save jwbowers/ab639e6a2045a054e457c52df03c0982 to your computer and use it in GitHub Desktop.
Save jwbowers/ab639e6a2045a054e457c52df03c0982 to your computer and use it in GitHub Desktop.
A simple sampling design for declare design
> pop <- declare_population(N,
+ covid=rbinom(N,size=1,prob=covidprob))
> samp <- declare_sampling(n=n)
> sim_design <- pop + samp
> set.seed(12345)
> tmp <- draw_data(redesign(sim_design,N=40,n=20,covidprob=.5))
> table(tmp$covid)
0 1
9 11
> stopifnot( abs( mean(tmp$covid) - .5 ) < .1)
> estimand <- declare_estimand(mean(covid),label="prop covid")
> estimator1 <- declare_estimator(covid~1,model=lm_robust, se_type="HC0", label="prop positive")
> estimator2 <- declare_estimator(,label="prop pos with error")
> sim_design_est <- sim_design + estimand + estimator1 ## + estimator2
> d_N40_n20_p5 <- redesign(sim_design_est,N=40,n=20,covidprob=.5)
> draw_estimands(d_N40_n20_p5)
estimand_label estimand
1 prop covid 0.6
> draw_estimands(d_N40_n20_p5)
estimand_label estimand
1 prop covid 0.5
> draw_estimands(d_N40_n20_p5)
estimand_label estimand
1 prop covid 0.4
> draw_estimates(d_N40_n20_p5)
estimator_label term estimate std.error statistic p.value conf.low conf.high df outcome
1 prop positive (Intercept) 0.4 0.1095445 3.651484 0.001697388 0.1707207 0.6292793 19 covid
>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment