Defines configuration parameters for the poll aggregator, including Stan settings, and election details.
Usage
configurar_agregador(
pesquisas = NULL,
resultado_eleicao_passada = NULL,
resultado_eleicao_atual = NULL,
historico_pesquisas = NULL,
candidaturas_1t = NULL,
candidaturas_2t = NULL,
direita_eleicao_atual = NULL,
direita_eleicao_passada = "Bolsonaro",
esquerda_eleicao_atual = NULL,
esquerda_eleicao_passada = "Lula",
eleicao_passada_primeiro_turno = "2/10/2022",
eleicao_passada_segundo_turno = "30/10/2022",
stan_cores = pmin(parallel::detectCores(), 4),
stan_chains = 4,
stan_warmup = 500,
stan_sampling = 500,
stan_init = 0.1,
stan_adapt_delta = 0.99,
saida_bases_tratadas = "resultados_agregador/bases_tratadas",
saida_modelos_brutos = "resultados_agregador/modelos_brutos"
)Arguments
- pesquisas
Path to a CSV file or URL containing current poll data. Defaults to a GitHub Raw URL.
- resultado_eleicao_passada
Path to a CSV file containing results from the previous election. Defaults to a GitHub Raw URL.
- resultado_eleicao_atual
Path to a CSV file containing results for the current election (useful for retrospective model). Defaults to a GitHub Raw URL.
- historico_pesquisas
Path to a CSV/RDS file containing historical poll data. If NULL (default), uses the package's internal dataset.
- candidaturas_1t
Character vector of candidates in the 1st round. If NULL, uses default candidates.
- candidaturas_2t
Character vector of candidates in the 2nd round. If NULL, uses default candidates.
- direita_eleicao_atual
Character vector of right-wing candidates in the current race. If NULL, uses default candidates. The model can compensate institute errors against right-wing candidates in the last election.
- direita_eleicao_passada
Name of the right-wing candidate in the previous election.
- esquerda_eleicao_atual
Character vector of left-wing candidates in the current race. If NULL, uses default candidates. The model can compensate institute errors against left-wing candidates in the last election.
- esquerda_eleicao_passada
Name of the left-wing candidate in the previous election.
- eleicao_passada_primeiro_turno
Date of the previous 1st round (e.g., "2/10/2022").
- eleicao_passada_segundo_turno
Date of the previous 2nd round (e.g., "30/10/2022").
- stan_cores
Number of CPU cores for Stan to use.
- stan_chains
Number of MCMC chains.
- stan_warmup
Number of warmup iterations per chain.
- stan_sampling
Number of sampling iterations per chain.
- stan_init
Initial value for Stan parameters.
- stan_adapt_delta
The target acceptance rate for Stan's NUTS algorithm.
- saida_bases_tratadas
Directory where treated data will be saved.
- saida_modelos_brutos
Directory where raw model objects will be saved.