BMRMM: Bayesian Markov Renewal Mixed Models (BMRMMs)

In BMRMM: An Implementation of the Bayesian Markov (Renewal) Mixed Models

Bayesian Markov Renewal Mixed Models (BMRMMs)

Description

Provides inference results of both transition probabilities and duration times using BMRMMs.

Usage

BMRMM(
  data,
  num.cov,
  cov.labels = NULL,
  state.labels = NULL,
  random.effect = TRUE,
  fixed.effect = TRUE,
  trans.cov.index = 1:num.cov,
  duration.cov.index = 1:num.cov,
  duration.distr = NULL,
  duration.incl.prev.state = TRUE,
  simsize = 10000,
  burnin = simsize/2
)

Arguments

data	a data frame containing – individual ID, covariate values, previous state, current state, duration times (if applicable), in that order.
num.cov	total number of covariates provided in data.
cov.labels	a list of vectors giving names of the covariate levels. Default is a list of numerical vectors.
state.labels	a vector giving names of the states. Default is a numerical vector.
random.effect	TRUE if population-level effects are considered. Default is TRUE.
fixed.effect	TRUE if individual-level effects are considered. Default is TRUE.
trans.cov.index	a numeric vector indicating the indices of covariates that are used for transition probabilities. Default is all of the covariates.
duration.cov.index	a numeric vector indicating the indices of covariates that are used for duration times. Default is all of the covariates.
duration.distr	a list of arguments indicating the distribution of duration times. Default is NULL, which is ignoring duration times.
duration.incl.prev.state	TRUE if the previous state is included in the inference of duration times. Default is TRUE.
simsize	total number of MCMC iterations. Default is 10000.
burnin	number of burn-ins for the MCMC iterations. Default is simsize/2.

Details

Users have the option to ignore duration times or model duration times as a discrete or continuous variable via defining duration.distr.

duration.distr can be one of the following:

• NULL: duration times are ignored. This is the default setting.

• list('mixgamma', shape, rate): duration times are modeled as a mixture gamma variable. shape and rate must be numeric vectors of the same length. The length indicates the number of mixture components.

• list('mixDirichlet', unit): duration times are modeled as a new state with discretization unit. The duration state is then analyzed along with the original states. For example, if an duration time entry is 20 and unit is 5, then the model will add 4 consecutive new states. If an duration time entry is 23.33 and unit is 5, then the model will still add 4 consecutive new states as the blocks are calculated with the floor operation.

Value

An object of class BMRMM consisting of results.trans and results.duration if duration times are analyzed as a continuous variable.

The field results.trans is a data frame giving the inference results of transition probabilities.

covs	covariates levels for each row of the data.
dpreds	maximum level for each related covariate.
MCMCparams	MCMC parameters including simsize, burnin and thinning factor.
tp.exgns.post.mean	posterior mean of transition probabilities for different combinations of covariates.
tp.exgns.post.std	posterior standard deviation of transition probabilities for different combinations of covariates.
tp.anmls.post.mean	posterior mean of transition probabilities for different individuals.
tp.anmls.post.std	posterior standard deviation of transition probabilities for different individuals.
tp.all.post.mean	posterior mean of transition probabilities for different combinations of covariates AND different individuals.
tp.exgns.diffs.store	difference in posterior mean of transition probabilities for every pair of covariate levels given levels of the other covariates.
tp.exgns.all.itns	population-level transition probabilities for every MCMC iteration.
clusters	number of clusters for each covariate for each MCMC iteration.
cluster_labels	the labels of the clusters for each covariate for each MCMC iteration.
type	a string identifier for results, which is "Transition Probabilities".
cov.labels	a list of string vectors giving labels of covariate levels.
state.labels	a list of strings giving labels of states.

The field results.duration is a data frame giving the inference results of duration times.

covs	covariates related to duration times.
dpreds	maximum level for each related covariate.
MCMCparams	MCMC parameters: simsize, burnin and thinning factor.
duration.times	duration times from the data set.
comp.assignment	mixture component assignment for each data point in the last MCMC iteration.
duration.exgns.store	posterior mean of mixture probabilities for different combinations of covariates of each MCMC iteration.
marginal.prob	estimated marginal mixture probabilities for each MCMC iteration.
shape.samples	estimated shape parameters for gamma mixtures for each MCMC iteration.
rate.samples	estimated rate parameters for gamma mixtures for each MCMC iteration.
clusters	number of clusters for each covariate for each MCMC iteration.
cluster_labels	the labels of the clusters for each covariate for each MCMC iteration.
type	a string identifier for results, which is "Duration Times".
cov.labels	a list of string vectors giving labels of covariate levels.

Author(s)

Yutong Wu, [e-mail]

Examples

# In the examples, we use a shorted version of the foxp2 dataset, foxp2sm

# ignores duration times and only models transition probabilities using all three covariates
results <- BMRMM(foxp2sm, num.cov = 2, simsize = 50)

# models duration times as a continuous variable with 3 gamma mixture components,
results <- BMRMM(foxp2sm, num.cov = 2, simsize = 50,
                 duration.distr = list('mixgamma', shape = rep(1,3), rate = rep(1,3)))

# models duration times as a discrete state with discretization 0.025 and
results <- BMRMM(foxp2sm, num.cov = 2, simsize = 50, 
                 duration.distr = list('mixDirichlet', unit = 0.025))

BMRMM documentation built on May 29, 2024, 3:53 a.m.