Description Usage Arguments Value
Function to run a Maximum-likelihood via Iterated Filtering algorithm.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 | mif(
ssm,
approx = c("ode", "sde", "psr"),
dt = NULL,
id = 0,
root = NULL,
n_parts = NULL,
iter = NULL,
n_obs = NULL,
cooling = NULL,
switch = NULL,
eps_switch = NULL,
eps_max = NULL,
eps_abs_integ = NULL,
eps_rel_integ = NULL,
like_min = NULL,
heat = NULL,
lag = NULL,
freeze_forcing = NULL,
interpolator = NULL,
verbose = FALSE,
warning = FALSE,
no_dem_sto = FALSE,
no_white_noise = FALSE,
no_diff = FALSE,
traj = TRUE,
trace = TRUE,
diag = TRUE,
prior = FALSE,
ic_only = FALSE,
seed_time = TRUE
)
|
ssm |
a |
approx |
character, approximation used to simulate
|
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 |
n_parts |
numeric, number of particles. |
iter |
numeric, number of iterations. |
n_obs |
numeric, number of observations to be fitted (for tempering). If |
cooling |
numeric, cooling factor (for sampling covariance live tuning or |
switch |
numeric, select switching iteration from initial covariance to empirical one ( |
eps_switch |
numeric, select number of burn-in iterations before tuning epsilon. |
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 |
heat |
numeric, re-heating across MIF iterations (scales standard deviation of proposals) |
lag |
numeric, lag for fixed-lag smoothing (proportion of the data) |
freeze_forcing |
character, freeze covariates to their value at specified date (in |
interpolator |
character, gsl interpolator for covariates |
verbose |
logical, print logs (verbose). Default to |
warning |
logical, print warnings. Default to |
no_dem_sto |
logical, turn off demographic stochasticity (if any). Default to |
no_white_noise |
logical, turn off white noises (if any). Default to |
no_diff |
logical, turn off diffusions (if any). Default to |
traj |
logical, print the trajectories. Default to |
trace |
logical, print the trace. Default to |
diag |
logical, print the diagnostics outputs (e.g. prediction residuals). Default to |
prior |
logical, add log(prior) to the estimated log-likelihood. Default to |
ic_only |
logical, only fit the initial condition using fixed lag smoothing. Default to |
seed_time |
logical, seed the random number generator with the current time. Default to |
a ssm
object updated with latest SSM output and ready to be piped into another SSM block.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.