Package 'walk'

The form of the rate matrix entries. Can be one of: "mult" for residency times movement, i.e., q_{ij} = q_r(i) * pi_m(i,j), "add" for a model of the form q_{ij} = q_r(i) + pi_m(i,j), and "sde" for a Langevin diffusion approximation. The "sde" is very similar to the additive model.

A link funtion for the residency or movement model. One of "soft_plus" or "log".

Scale parameter for the "soft_plus" link function. Ignored for link="log". The a parameter determines the approximation to a hard plus function, i.e., as a becomes large the soft plus function converges to g^{-1}(x) = max(0,x). For this specification, a must be greater than or equal to 1.

Lower bound for general logit link

Upper bound for general logit link

A named list for the residency model. Must contain elements: form, link, and a. The easiest way to construct this is the ctmc_model function.

A named list for the movement portion of the model. See q_r.

Logical. Should a zero-inflation parameter be used. Necessary when there might be outliers when the location likelihood does not overlap likely locations.

Vector of probabilities for the initial location. Can be one of "uniform" for a uniform distribution over all cells, "stationary" to use the stationary distribution implied by the parameters, or a numeric vector the same length as the number of cells.

Should the movement portion be adjusted to sum to 1. This is the parameterization suggested by Hewitt et al. (2023).

Positive value for the maximum movement rate for all cells. All cells with modeled total movement rates above this value will be truncated to clip.

SpatRaster object

SpatRaster object

Parameter vector

List of required data objects to evaluate likelihood

For developers only, leave in the default setting.

Extra wiggle room for ignored arguments.

An R formula object for the paramter vector.

Design data list

Fraction of time unit in processed telemetry data

Cummulative probability of residency time

The maximum value for the clip value

A design data list produced by the function make_walk_data.

A named list giving specifications for the model. See ctmc_control.

An optional penalty function. Should be on the scale of a log-prior distribution.

Logical. Should the Hessian matrix be calculated to obtain the parameter variance-covariance matrix.

Calculate real values for expected residency, cell transition probabilities, and outlier proportion for observations.

Optional starting values for the parameter must be a list of the form list(beta_l=c(), beta_q_r=c(), beta_q_r=c()).

Optimization method. See optimr

Logical. Should the likelihood be optimized?

Error rate of matrix exponential calculation. Defaults to 1.0e-8. This is a generous value. If the model is running slow, you can try reducing it to, say, 1.0e-4.

Integer from 1-4. Opens browser() at various points in the function call. Mostly for package developers.

Additional arguments passed to the optimization function optimr from the optimx-package.

numeric value

Lower bound

Upper bound

Model estimated parameters

Optional variance-covariance matrix for the parameters. If none is given no standard errors will be provided for the real parameter estimates

The data list from a fitted model object

SpatRaster object

Logical, Should the gradient be scaled by resolution of the coordinates. Defaults to scale = TRUE

A fitted model object from fit_ctmc.

A vector of probabilities. Will return columns with highest probability area for each specified probability. E.g., hpd=c(0.5, 0.95) will return 2 extra columns with 50 and 95% HPD densities.

Method used for eigen decomposition. One of "lu" or "arpack".

Extra arguments to pass to eigs

A fitted model object from fit_ctmc.

Logical. Should the matrix be returned in a sparse format from the Matrix package. Defaults to sparse = TRUE.

A sparse matrix with rows corresponding to time steps and columns corresponding to cells. The entries are probabilities that the animal is located in the corresponding cell and time.

A SpatRaster object from the terra package.

A character vector of names of cell_data layers for which gradient covariates will be constructed.

Raster mask for inaccessible cells when cell_data is of type SpatRaster from the terra package. This is ignored if cell_data is an POLYGON data frame from the sf package.

Neighborhood structure, one of "rook" or "queen".

Debugging level: 1-3 mainly for package developers.

Ignored arguments.

A fitted model object produced by fit_ctmc

A design data list produced by the function make_walk_data.

Additional POSIX times for which location prediction is desired.

Level for truncating small location prediction probabilities.

Developer debugging.

Aditional arguments passed to internal function ctmc_predict_arma.

A telemetry object from the ctmm package

A ⁠\link[terra]{SpatRaster}⁠ stack of covariates that will be used in CTMC movement modeling. Cells with NA values will be considered areas to the animal cannot travel, i.e., likelihood surfaces will be 0 for those cells.

Unit of measurement for numeric conversion of the POSIX timestamps in the telemetry data. Defaults to "hours". Can specify, e.g., "seconds" or "days"

Type of object returned. One if "data.frame", "sparse" (sparse matrix), "matrix_df" (matrix form of "data.frame"), or "dense" (dense matrix).

The maximum error in meters. If unspecified it will be set to 4 times the maximum error standard deviation as determined by the UERE and HDOP of the telemetry data.

The smallest probability value that is considered to be > 0. Defaults to 1.0e-8.

Logical. If there are locations outside the admissible areas of movement return the observations and raster cell numbers.

SpatRaster object

A spatRaster object from the terra package

a polygon object that denotes places where the animal cannot travel

A SpatRaster where cells will be set to NA that are inside the polygon object.

The amount of overlap betwen the raster cell and polygon such that the cell is considered "inside" the polygon. Defaults to overlap = 0.95. If overlap = 1 then a cell has to be completely inside the polygon to be masked.

numeric value

scale parameter, a must be >1.