smc: Run a SMC
In StateSpaceModels/ssminr: ssminr

Function to run a Sequential Monte-Carlo algorithm on a ssm

smc(
  ssm,
  approx = c("ode", "sde", "psr"),
  dt = NULL,
  id = 0,
  root = NULL,
  n_parts = NULL,
  n_thread = 1,
  iter = NULL,
  n_obs = NULL,
  eps_max = NULL,
  eps_abs_integ = NULL,
  eps_rel_integ = NULL,
  like_min = NULL,
  freeze_forcing = NULL,
  interpolator = NULL,
  verbose = FALSE,
  warning = FALSE,
  no_dem_sto = FALSE,
  no_white_noise = FALSE,
  no_diff = FALSE,
  traj = TRUE,
  hat = FALSE,
  trace = TRUE,
  diag = TRUE,
  prior = FALSE,
  no_filter = FALSE,
  seed_time = TRUE
)

`ssm`	a `ssm` object, returned by `new_ssm`.
`approx`	character, approximation used to simulate `ssm`: `"ode"` deterministic approximation, based on Ordinary Differential Equations. `"sde"` diffusion approximation, based on Stochastic Differential Equations. `"psr"` Euler-multinomial approximation, based on Poisson process with Stochastic Rates.
`dt`	numeric, integration time step.
`id`	integer, unique integer identifier that will be appended to the output.
`root`	character, root path for output files (if any) (no trailing slash). If `NULL` (default), outputs are written in "your_model_path/the_name_of_the_wrapper".
`n_parts`	numeric, number of particles.
`n_thread`	numeric, number of threads to be used. Default to 1. Use `"max"` to set it automatically to the number of cores available on the machine using `detectCores`.
`iter`	numeric, number of iterations.
`n_obs`	numeric, number of observations to be fitted (for tempering). If `NULL` (default), all observations are fitted.
`eps_max`	numeric, maximum value allowed for epsilon.
`eps_abs_integ`	numeric, absolute error for adaptive step-size control.
`eps_rel_integ`	numeric, relative error for adaptive step-size control.
`like_min`	numeric, particles with likelihood smaller than `like_min` are considered lost. If `NULL` (default) lower bound on likelihood based on machine precision.
`freeze_forcing`	character, freeze covariates to their value at specified date (in `YYYY-MM-DD` format).
`interpolator`	character, gsl interpolator for covariates
`verbose`	logical, print logs (verbose). Default to `FALSE`.
`warning`	logical, print warnings. Default to `FALSE`.
`no_dem_sto`	logical, turn off demographic stochasticity (if any). Default to `FALSE`.
`no_white_noise`	logical, turn off white noises (if any). Default to `FALSE`.
`no_diff`	logical, turn off diffusions (if any). Default to `FALSE`.
`traj`	logical, print the trajectories. Default to `TRUE`.
`hat`	logical, print the state estimates. Default to `FALSE`.
`trace`	logical, print the trace. Default to `TRUE`.
`diag`	logical, print the diagnostics outputs (e.g. prediction residuals). Default to `TRUE`.
`prior`	logical, add log(prior) to the estimated log-likelihood. Default to `TRUE`.
`no_filter`	logical, do not filter. Default to `FALSE`.
`seed_time`	logical, seed the random number generator with the current time. Default to `TRUE`.