検査数が3桁になって推定有病率が減ってきたな。

> subjects=c(74,95,87,143,244,330)
> positives=c(17,41,47,40,63,68)
> PCRs3(subjects,positives,iter=10000,warmup=1000)
Inference for Stan model: corona31.
4 chains, each with iter=10000; warmup=1000; thin=1;
post-warmup draws per chain=9000, total post-warmup draws=36000.

mean se_mean sd 2.5% 50% 97.5% n_eff Rhat
prev 0.37289 0.00136 0.13285 0.12790 0.36264 0.67656 9545 1.00005
sen 0.56775 0.00095 0.10235 0.36708 0.56882 0.76252 11488 1.00028
spc 0.87547 0.00045 0.05056 0.76198 0.88107 0.95746 12707 1.00023
p 0.28424 0.00008 0.01441 0.25667 0.28411 0.31302 34879 1.00001
lp__ -614.08170 0.01429 1.32344 -617.51076 -613.72776 -612.56237 8576 1.00036

Samples were drawn using NUTS(diag_e) at Mon Mar 30 21:01:47 2020.
For each parameter, n_eff is a crude measure of effective sample size,
and Rhat is the potential scale reduction factor on split chains (at
convergence, Rhat=1).
mean lower upper
0.37288732 0.09822213 0.63719043