sample_posterior() draws samples from the
posterior distribution of the specified model given a data set with
visit data.
sample_prior() draws samples from the
prior distribution of the specified model object.
Arguments
- model
an object of class srpmodel containing prior information
- data
a data frame with variables
subject_id<chr>(subject identifier),group_id<chr>(group identifier),t<dbl>(time of visit, relative to first visit in study),state<chr>(state recorded at visit). Allowed states are "stable", "response", "progression" (or death), and "EOF" (end of follow-up). The EOF state marks the end of an individual's follow-up before the absorbing state "progression".- now
numeric, time from first visit in data if different form last recorded visit
- nsim
integer, number of samples to draw
- seed
integer, fixed random seed; NULL for no fixed seed
- warmup
integer, number of warm-up samples for the MCMC sampler before retaining samples; see
warmupparameter inrstan::stan().- nuts_control
list, parameters for NUTS algorithm see
controlargument inrstan::stan()- acceptable_divergent_transition_fraction,
numeric between 0 and 1 giving the acceptable fraction of divergent transitions before throwing an error
- ...
further arguments passed to method implementations
Value
A rstan::stanfit object with posterior samples.
Examples
mdl <- create_srpmodel(A = define_srp_prior())
tbl <- tibble::tibble(
subject_id = c("A1", "A1"),
group_id = c("A", "A"),
t = c(0, 1.5),
state = c("stable", "response")
)
sample_posterior(mdl, tbl, seed = 42L)
#> Inference for Stan model: srp_model_simple.
#> 1 chains, each with iter=2500; warmup=500; thin=1;
#> post-warmup draws per chain=2000, total post-warmup draws=2000.
#>
#> mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat
#> p[1] 0.62 0.00 0.22 0.16 0.47 0.64 0.81 0.97 2190 1
#> median_t[1,1] 3.72 0.23 7.51 0.42 1.10 1.95 3.80 17.83 1101 1
#> median_t[1,2] 16.23 0.72 26.85 0.71 3.32 7.80 17.96 81.91 1404 1
#> median_t[1,3] 17.67 0.79 28.91 1.02 3.55 8.00 18.84 90.19 1325 1
#> shape[1,1] 1.54 0.01 0.47 0.83 1.21 1.47 1.81 2.66 2493 1
#> shape[1,2] 1.57 0.01 0.49 0.83 1.22 1.50 1.82 2.68 3233 1
#> shape[1,3] 1.60 0.01 0.51 0.85 1.22 1.52 1.91 2.73 3183 1
#> scale[1,1] 5.01 0.32 10.51 0.53 1.40 2.48 5.02 25.33 1066 1
#> scale[1,2] 21.09 0.95 35.37 0.90 4.25 10.13 23.37 103.49 1385 1
#> scale[1,3] 22.94 1.02 38.15 1.26 4.57 10.34 24.44 119.57 1409 1
#> lp__ -8.45 0.08 2.00 -13.16 -9.47 -8.11 -6.99 -5.59 660 1
#>
#> Samples were drawn using NUTS(diag_e) at Mon Apr 17 14:49:45 2023.
#> 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).
sample_prior(mdl, seed = 42L)
#> Inference for Stan model: srp_model_simple.
#> 1 chains, each with iter=2500; warmup=500; thin=1;
#> post-warmup draws per chain=2000, total post-warmup draws=2000.
#>
#> mean se_mean sd 2.5% 25% 50% 75% 97.5% n_eff Rhat
#> p[1] 0.50 0.00 0.24 0.06 0.31 0.50 0.68 0.93 4210 1
#> median_t[1,1] 18.44 1.66 60.24 0.66 3.35 7.62 17.93 93.39 1313 1
#> median_t[1,2] 15.94 0.70 27.92 0.71 3.45 7.75 17.83 77.51 1570 1
#> median_t[1,3] 16.69 0.76 31.94 0.68 3.35 7.63 17.93 93.79 1761 1
#> shape[1,1] 1.57 0.01 0.49 0.83 1.21 1.51 1.86 2.70 2917 1
#> shape[1,2] 1.56 0.01 0.49 0.84 1.20 1.48 1.85 2.68 4107 1
#> shape[1,3] 1.58 0.01 0.52 0.83 1.19 1.51 1.86 2.80 3429 1
#> scale[1,1] 23.99 2.22 80.73 0.86 4.36 9.84 23.33 123.84 1321 1
#> scale[1,2] 20.70 0.90 35.96 0.94 4.45 9.97 22.84 100.43 1609 1
#> scale[1,3] 21.89 1.05 43.80 0.84 4.33 9.75 22.90 120.76 1735 1
#> lp__ -6.02 0.07 1.97 -10.54 -7.08 -5.65 -4.56 -3.28 768 1
#>
#> Samples were drawn using NUTS(diag_e) at Mon Apr 17 14:49:45 2023.
#> 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).