Modify simulation specifications — sim

Simulation specifications are the settings that determine how the model is simulated, such as the integration method, start and stop time, and timestep. Modify these specifications for an existing stock-and-flow model.

Usage

sim_specs(
  sfm,
  method = "euler",
  start = "0.0",
  stop = "100.0",
  dt = "0.01",
  saveat = dt,
  seed = NULL,
  time_units = "s",
  language = "R",
  ...
)

Arguments

sfm: Stock-and-flow model, object of class sdbuildR_xmile.
method: Integration method. Defaults to "euler".
start: Start time of simulation. Defaults to 0.
stop: End time of simulation. Defaults to 100.
dt: Timestep of solver. Defaults to 0.01.
saveat: Timestep at which to save computed values. Defaults to dt.
seed: Seed number to ensure reproducibility across runs in case of random elements. Must be an integer. Defaults to NULL (no seed).
time_units: Simulation time unit, e.g. 's' (second). Defaults to "s".
language: Coding language in which to simulate model. Either "R" or "Julia". Julia is necessary for using units or delay functions. Defaults to "R".
...: Optional additional parameters

Value

Updated stock-and-flow model with new simulation specifications

Examples

sfm = xmile("predator-prey") %>%
sim_specs(start = 1960, stop = 2010, dt = 0.1)
sim = simulate(sfm)
plot(sim)


# Change the simulation method to "rk4"
sfm = sfm %>% sim_specs(method = "rk4")

# Change the time units to "years", such that one time unit is one year
sfm = sfm %>% sim_specs(time_units = "years")

# To save storage but not affect accuracy, use saveat
sfm = sfm %>% sim_specs(saveat = 0.1, dt = 0.001)
sim = simulate(sfm)
head(as.data.frame(sim))
#>     time predator     prey predator_births predator_deaths prey_births
#> 1 1960.0 10.00000 50.00000        12.50000        5.000000    25.00000
#> 2 1960.1 10.77798 49.90401        13.44661        5.388992    24.95200
#> 3 1960.2 11.61083 49.60798        14.39975        5.805415    24.80399
#> 4 1960.3 12.49553 49.10233        15.33899        6.247763    24.55116
#> 5 1960.4 13.42704 48.38153        16.24052        6.713520    24.19077
#> 6 1960.5 14.39813 47.44494        17.07796        7.199065    23.72247
#>   prey_deaths
#> 1    25.00000
#> 2    26.89323
#> 3    28.79949
#> 4    30.67797
#> 5    32.48104
#> 6    34.15592

# Specify seed for reproducibility
sfm = sfm %>% sim_specs(seed = 1) %>%
 # Add stochastic initial condition
 build(c("predator", "prey"), eqn = "runif(1, 20, 50)")
sim1 = simulate(sfm)
sim2 = simulate(sfm)
plot(sim1)

plot(sim2)


# Removing the seed yields variation
sfm = sfm %>% sim_specs(seed = NULL)
sim1 = simulate(sfm)
sim2 = simulate(sfm)
plot(sim1)

plot(sim2)


# Change the simulation language to Julia to use units and delay functions
sfm = sfm %>% sim_specs(language = "Julia")