sample_pum_dynamic {pumBayes} | R Documentation |
Generate posterior samples from the dynamic probit unfolding model
Description
This function generates posterior samples for all parameters based on the dynamic probit unfolding model.
Usage
sample_pum_dynamic(
vote_info,
years_v,
hyperparams,
control,
sign_refs,
verbose = FALSE,
pre_run = NULL,
appended = FALSE
)
Arguments
vote_info |
A logical vote matrix where rows represent members and columns represent issues. The entries should be FALSE ("No"), TRUE ("Yes"), or NA (missing data). |
years_v |
A vector representing the time period for each vote in the model. |
hyperparams |
A list of hyperparameter values including: - 'beta_mean': Prior mean of beta. - 'beta_var': Prior variance of beta. - 'alpha_mean': A vector of 2 values for the prior means of alpha1 and alpha2. - 'alpha_scale': Scale parameter for alpha1 and alpha2. - 'delta_mean': A vector of 2 values for the prior means of delta1 and delta2. - 'delta_scale': Scale parameter for delta1 and delta2. - 'rho_mean': Prior mean of the autocorrelation parameter 'rho'. - 'rho_sigma': Standard deviation of the prior for 'rho'. |
control |
A list specifying the MCMC configurations, including: - 'num_iter': Total number of iterations. - 'burn_in': The number of initial iterations to discard as part of the burn-in period before retaining samples. - 'keep_iter': Interval at which samples are retained. - 'flip_rate': Probability of directly flipping signs in the M-H step, rather than resampling from the priors. - 'sd_prop_rho': Proposal standard deviation for 'rho' in the Metropolis-Hastings step. |
sign_refs |
A list containing sign constraints, including: - 'pos_inds': Indices of members constrained to have positive values. - 'neg_inds': Indices of members constrained to have negative values. - 'pos_year_inds': List of years corresponding to each 'pos_ind'. - 'neg_year_inds': List of years corresponding to each 'neg_ind'. |
verbose |
Logical. If 'TRUE', prints progress and additional information during the sampling process. |
pre_run |
A list containing the output from a previous run of the function. If provided, the last iteration of the previous run will be used as the initial point of the new run. Defaults to 'NULL'. |
appended |
Logical. If 'TRUE', the new samples will be appended to the samples from the previous run. Defaults to 'FALSE'. |
Value
A list containing: - 'beta': A data frame of posterior samples for beta. - 'alpha1': A data frame of posterior samples for alpha1. - 'alpha2': A data frame of posterior samples for alpha2. - 'delta1': A data frame of posterior samples for delta1. - 'delta2': A data frame of posterior samples for delta2. - 'rho': A data frame of posterior samples for rho.
Examples
# Long-running example
data(scotus.1937.2021)
hyperparams = list(alpha_mean = c(0, 0), alpha_scale = 5,
delta_mean = c(-2, 10), delta_scale = sqrt(10),
rho_mean = 0.9, rho_sigma = 0.04)
control = list(num_iter = 2, burn_in = 0, keep_iter = 1, flip_rate = 0.1, sd_prop_rho = 0.1)
sign_refs = list(pos_inds = c(39, 5), neg_inds = c(12, 29),
pos_year_inds = list(1:31, 1), neg_year_inds = list(1:29, 1:24))
scotus.pum = sample_pum_dynamic(mqVotes, mqTime, hyperparams, control, sign_refs,
verbose = FALSE, pre_run = NULL, appended = FALSE)