calc_step_lik: Step likelihood.

In a-parton/CTStepTurn: Fitting the continuous-time step-and-turn movement model

Description

calc_step_lik returns the likelihood of a set of steps given speed parameters.

Usage

calc_step_lik(fixed_constant, refined_path, move_params, behav_params)

Arguments

fixed_constant	List with components: num_states Numeric, number of behavioural states. indep_step Boolean, True if model is independent step, False if it is correlated step. q.speed_mean Vector, index positions within move_params relating to speed mean parameters. q.speed_corr Vector, index positions within move_params relating to speed correlation parameters. q.speed_var Vector, index positions within move_params relating to speed variance parameters.
refined_path	List with components: time_diffs Vector, differences in the refined time scale. behavs Vector, refined behavioural/state sequence. steps Vector, refined step process.
move_params	Vector of movement parameters (as turn_var_1, turn_var_2, mean_speed_1, .., speed_var_2).
behav_params	List with components: lambda Vector (length num_states) of switching rates. q Matrix (square size num_states with NA diagonal elements) of switching probabilities. Note: row = from, column = to.

Value

Numeric, likelihood of the set of steps.

See Also

Other Speed parameters: calc_cond_step_dist, calc_markov_steady_state, update_speed_param

Examples

const <- list(num_states = 3,
              indep_step = FALSE,
              q.speed_mean = 4:6,
              q.speed_corr = 7:9,
              q.speed_var = 10:12)
path <- list(steps = c(5.3, 6.8, 15.3, 14.5, 9.4, 3.4, 8.5, 14.6, 17.8, 22.4, 16.7),
             time_diffs = c(1, 1, 1.5, 1.5, 1, 0.5, 1, 1, 1, 2, 1),
             behavs = c(1, 1, 1, 2, 2, 3, 3, 2, 2, 2, 1))
move_params <- c(NA, NA, NA, 5, 10, 3, 0.1, 0.2, 0.3, 4, 6, 2)
behav_params <- list(lambda = c(0.1, 0.2, 0.3),
                     q = matrix(c(NA, 0.5, 0.2, 0.3, NA, 0.8, 0.7, 0.5, NA), nrow = 3))

calc_step_lik(const, path, move_params, behav_params)
# [1] -78.24704

path <- list(steps = c(5.3, 6.8, 15.3, 14.5, 9.4, 3.4, 8.5, 14.6, 17.8, 22.4, 16.7),
             time_diffs = c(2, 2, 0.5, 1, 2, 0.25, 2, 1.5, 1, 1.2, 0.6),
             behavs = c(1, 1, 1, 2, 2, 3, 3, 2, 2, 2, 1))
calc_step_lik(const, path, move_params, behav_params)
# [1] -791.1856

a-parton/CTStepTurn documentation built on May 14, 2019, 4:17 p.m.