mobility: Fit mobility model to data

In COVID-19-Mobility-Data-Network/mobility: Tools for Analyzing Human Mobility Data

Description

This function fits a variety of mobility models to a supplied movement matrix (M) and covariates (D and N) using Bayesian MCMC inference. The function specifies the type of mobility model and serves as a wrapper for the fit_jags function.

Usage

mobility(
  data,
  model,
  type,
  hierarchical = FALSE,
  n_chain = 2,
  n_burn = 1000,
  n_samp = 1000,
  n_thin = 1,
  DIC = FALSE,
  parallel = FALSE
)

Arguments

data	a list containing mobility data and covariates. The list object must include EITHER M, D, and N OR M, D, N_orig, and N_dest. Details about each item are as follows: M named matrix of trip counts between all ij location pairs D named matrix of distances between all ij location pairs N named vector of population sizes for all locations (either N or both N_orig and N_dest must be supplied) N_orig named vector of population sizes for each origin N_dest named vector of population sizes for each destination
model	character vector indicating which mobility model to fit to the data. Mobility models include: 'gravity', 'radiation', and 'departure-diffusion'.
type	character vector indicating the particular sub-type of mobility model to fit. Gravity model types include: basic, transport, power, exp, power_norm, exp_norm, and Marshall Radiation model types include: basic and finite Departure-diffusion model types include: power, exp, and radiation See model list vignette for more detailed description of each model type.
hierarchical	Applies only to the 'departure-diffusion' model. A logical argument indicating whether or not to fit τ as a single parameter or hierarchically (as τ_{pop} and τ_i).
n_chain	number of MCMC sampling chains
n_burn	number of iterations to discard before sampling of chains begins (burn in)
n_samp	number of iterations to sample each chain
n_thin	interval to thin samples
DIC	logical indicating whether or not to calculate the Deviance Information Criterion (DIC) (default = FALSE)
parallel	logical indicating whether or not to run MCMC chains in parallel or sequentially (default = FALSE)

Value

An object of class mobility.model containing model information, data, and fitted model parameters

Author(s)

John Giles

See Also

Other model: check(), compare(), fit_jags(), fit_prob_travel(), predict(), residuals(), summary()

Examples

mod <- mobility(data=mobility_matrices, model='gravity', type='transport')

mod <- mobility(data=mobility_matrices, model='radiation', type='finite')

mod <- mobility(data=mobility_matrices, model='departure-diffusion', type='power')

COVID-19-Mobility-Data-Network/mobility documentation built on Nov. 22, 2021, 12:17 a.m.