Skip to contents

Fit continuous-time movement models

Source: R/fct_sims.R
fitting_model.Rd

This function fits continuous-time movement models to simulated location data using the ctmm package. It estimates movement parameters for each simulated trajectory, optionally running in parallel for efficiency. It supports both home range and speed estimation workflows.

Usage

fitting_model(obj, set_target = c("hr", "ctsd"), ...)

Arguments

obj

A list of simulated movement datasets, each formatted as a telemetry object compatible with ctmm.

set_target

A character vector specifying the research goals. Options include:

  • "hr" — Home range estimation.

  • "ctsd" — Speed and distance estimation.

...

Optional control parameters passed via .... These include .dur, .dti, .tau_p, .tau_v, .error_m, .check_sampling, .rerun, .parallel, and .trace. See Details for their descriptions.

Value

A list of fitted movement models (class ctmm), one per simulation. Each model is recentered to the origin (x = 0, y = 0).

Details

The function generates initial parameter estimates for each dataset using ctmm::ctmm.guess(). If the data includes simulated location error, it uses an error model accordingly. When .check_sampling is TRUE, it compares the sampling duration and interval against optimal thresholds derived from the provided autocorrelation timescales. Models are fitted using ctmm::ctmm.select(), which performs model selection to find the best-fit movement process. If .rerun is enabled, the function identifies simulations with effective sample sizes below 0.1 and attempts to reselect models for those. Finally, all fitted models are recentered to (0, 0) for downstream consistency.

The following arguments can be supplied via ...:

  • .dur: A list with elements value (numeric) and unit (string), specifying the maximum study duration. Example: list(value = 2, unit = "months").

  • .dti: A list with elements value (numeric) and unit (string), specifying the intended sampling interval. Example: list(value = 1, unit = "day").

  • .tau_p: A list of position autocorrelation timescales. Optional, but required if .check_sampling = TRUE.

  • .tau_v: A list of velocity autocorrelation timescales. Optional, but required if .check_sampling = TRUE.

  • .error_m: A numeric value specifying location error in meters (used for simulation).

  • .check_sampling: Logical; if TRUE, checks whether the sampling schedule meets minimum requirements for reliable model fitting via ctmm::ctmm.fit(). This feature is experimental and may change in future versions.

  • .rerun: Logical; if TRUE, re-runs model selection when simulations result in very low effective sample sizes, to avoid convergence issues.

  • .parallel: Logical; if TRUE, enables parallel computation.

  • .trace: Logical; if TRUE, print progress and timing messages to the console.

Note

This function is intended for internal use and may assume inputs follow specific structure and constraints not referenced explicitly.

See also

ctmm::ctmm.guess(), ctmm::ctmm.select()