R/SEIR_outbreak_model_fitting.R
seir_model_fit.RdCreate an instance of the hierarchical SEIR Stan model incorporating various data elements and sample model.
seir_model_fit( stan_model = NULL, tmax, n_outbreaks, outbreak_cases, outbreak_sizes, intervention_switch = TRUE, multilevel_intervention = FALSE, priors = prior_list, chains = 4, iter = 600, seed = 42, fit_type = "NUTS", data_model = "poisson" )
| stan_model | [rstan] model object |
|---|---|
| tmax | Total number of time-points in observation |
| n_outbreaks | Total number of outbreaks |
| outbreak_cases | Number of daily reported cases by outbreak |
| outbreak_sizes | The total size of each facility (initial number of susceptible and exposed) |
| intervention_switch | Describes whether interventions occur in data (default TRUE) |
| multilevel_intervention | Describes whether intervention occurs |
| chains | Number of chains to sample |
| iter | number of iterations of MCMC |
| seed | The seed for random number generation. Set to replicate results. |
| fit_type | string "NUTS" or "VB" (VB quicker but less accurate). |
| data_model | string "poisson" or "negative_binomial". If negative_binomial selected then uses phi prior to control for overdispersion |
| prior_list | List of priors. See [prior_list] |
An object of class `stanfit` returned by sampling
Mike Irvine
stan_mod <- rstan::stan_model(system.file("stan", "hierarchical_SEIR_incidence_model.stan", package = "cr0eso")) tmax <- 5 pop_size <- 100 dim(pop_size) <- c(1) example_incidence <- matrix(c(1,1,2,3,2),ncol=1) fit <- seir_model_fit(stan_model = stan_mod, tmax,1,example_incidence,pop_size)#> #> SAMPLING FOR MODEL 'hierarchical_SEIR_incidence_model' NOW (CHAIN 1). #> Chain 1: #> Chain 1: Gradient evaluation took 0 seconds #> Chain 1: 1000 transitions using 10 leapfrog steps per transition would take 0 seconds. #> Chain 1: Adjust your expectations accordingly! #> Chain 1: #> Chain 1: #> Chain 1: Iteration: 1 / 600 [ 0%] (Warmup) #> Chain 1: Iteration: 60 / 600 [ 10%] (Warmup) #> Chain 1: Iteration: 120 / 600 [ 20%] (Warmup) #> Chain 1: Iteration: 180 / 600 [ 30%] (Warmup) #> Chain 1: Iteration: 240 / 600 [ 40%] (Warmup) #> Chain 1: Iteration: 300 / 600 [ 50%] (Warmup) #> Chain 1: Iteration: 301 / 600 [ 50%] (Sampling) #> Chain 1: Iteration: 360 / 600 [ 60%] (Sampling) #> Chain 1: Iteration: 420 / 600 [ 70%] (Sampling) #> Chain 1: Iteration: 480 / 600 [ 80%] (Sampling) #> Chain 1: Iteration: 540 / 600 [ 90%] (Sampling) #> Chain 1: Iteration: 600 / 600 [100%] (Sampling) #> Chain 1: #> Chain 1: Elapsed Time: 1.674 seconds (Warm-up) #> Chain 1: 1.025 seconds (Sampling) #> Chain 1: 2.699 seconds (Total) #> Chain 1: #> #> SAMPLING FOR MODEL 'hierarchical_SEIR_incidence_model' NOW (CHAIN 2). #> Chain 2: #> Chain 2: Gradient evaluation took 0 seconds #> Chain 2: 1000 transitions using 10 leapfrog steps per transition would take 0 seconds. #> Chain 2: Adjust your expectations accordingly! #> Chain 2: #> Chain 2: #> Chain 2: Iteration: 1 / 600 [ 0%] (Warmup) #> Chain 2: Iteration: 60 / 600 [ 10%] (Warmup) #> Chain 2: Iteration: 120 / 600 [ 20%] (Warmup) #> Chain 2: Iteration: 180 / 600 [ 30%] (Warmup) #> Chain 2: Iteration: 240 / 600 [ 40%] (Warmup) #> Chain 2: Iteration: 300 / 600 [ 50%] (Warmup) #> Chain 2: Iteration: 301 / 600 [ 50%] (Sampling) #> Chain 2: Iteration: 360 / 600 [ 60%] (Sampling) #> Chain 2: Iteration: 420 / 600 [ 70%] (Sampling) #> Chain 2: Iteration: 480 / 600 [ 80%] (Sampling) #> Chain 2: Iteration: 540 / 600 [ 90%] (Sampling) #> Chain 2: Iteration: 600 / 600 [100%] (Sampling) #> Chain 2: #> Chain 2: Elapsed Time: 1.567 seconds (Warm-up) #> Chain 2: 1.09 seconds (Sampling) #> Chain 2: 2.657 seconds (Total) #> Chain 2: #> #> SAMPLING FOR MODEL 'hierarchical_SEIR_incidence_model' NOW (CHAIN 3). #> Chain 3: #> Chain 3: Gradient evaluation took 0 seconds #> Chain 3: 1000 transitions using 10 leapfrog steps per transition would take 0 seconds. #> Chain 3: Adjust your expectations accordingly! #> Chain 3: #> Chain 3: #> Chain 3: Iteration: 1 / 600 [ 0%] (Warmup) #> Chain 3: Iteration: 60 / 600 [ 10%] (Warmup) #> Chain 3: Iteration: 120 / 600 [ 20%] (Warmup) #> Chain 3: Iteration: 180 / 600 [ 30%] (Warmup) #> Chain 3: Iteration: 240 / 600 [ 40%] (Warmup) #> Chain 3: Iteration: 300 / 600 [ 50%] (Warmup) #> Chain 3: Iteration: 301 / 600 [ 50%] (Sampling) #> Chain 3: Iteration: 360 / 600 [ 60%] (Sampling) #> Chain 3: Iteration: 420 / 600 [ 70%] (Sampling) #> Chain 3: Iteration: 480 / 600 [ 80%] (Sampling) #> Chain 3: Iteration: 540 / 600 [ 90%] (Sampling) #> Chain 3: Iteration: 600 / 600 [100%] (Sampling) #> Chain 3: #> Chain 3: Elapsed Time: 11.104 seconds (Warm-up) #> Chain 3: 0.705 seconds (Sampling) #> Chain 3: 11.809 seconds (Total) #> Chain 3: #> #> SAMPLING FOR MODEL 'hierarchical_SEIR_incidence_model' NOW (CHAIN 4). #> Chain 4: #> Chain 4: Gradient evaluation took 0 seconds #> Chain 4: 1000 transitions using 10 leapfrog steps per transition would take 0 seconds. #> Chain 4: Adjust your expectations accordingly! #> Chain 4: #> Chain 4: #> Chain 4: Iteration: 1 / 600 [ 0%] (Warmup) #> Chain 4: Iteration: 60 / 600 [ 10%] (Warmup) #> Chain 4: Iteration: 120 / 600 [ 20%] (Warmup) #> Chain 4: Iteration: 180 / 600 [ 30%] (Warmup) #> Chain 4: Iteration: 240 / 600 [ 40%] (Warmup) #> Chain 4: Iteration: 300 / 600 [ 50%] (Warmup) #> Chain 4: Iteration: 301 / 600 [ 50%] (Sampling) #> Chain 4: Iteration: 360 / 600 [ 60%] (Sampling) #> Chain 4: Iteration: 420 / 600 [ 70%] (Sampling) #> Chain 4: Iteration: 480 / 600 [ 80%] (Sampling) #> Chain 4: Iteration: 540 / 600 [ 90%] (Sampling) #> Chain 4: Iteration: 600 / 600 [100%] (Sampling) #> Chain 4: #> Chain 4: Elapsed Time: 12.309 seconds (Warm-up) #> Chain 4: 0.706 seconds (Sampling) #> Chain 4: 13.015 seconds (Total) #> Chain 4:#> Warning: Tail Effective Samples Size (ESS) is too low, indicating posterior variances and tail quantiles may be unreliable. #> Running the chains for more iterations may help. See #> http://mc-stan.org/misc/warnings.html#tail-ess