gibbs_ad_use: Gibbs Sampler for Archaeological Dates: Artifact Use

In eratosthenes: Archaeological Synchronism

Gibbs Sampler for Archaeological Dates: Artifact Use

Description

Using the results of gibbs_ad, estimate a single density for the date of use of an artifact or artifact type. Multiple artifacts and types can be given, which will be pooled into a single type. For example, one can input several individual finds via their id number as comprising a type, or multiple (sub)types/classes as a single type, (e.g., "MGS V amphora", "MGS VI amphora", "MGS V/VI amphora" to construct one group).

Usage

gibbs_ad_use(
  marginalized,
  finds,
  id = NULL,
  type = NULL,
  type_name = NULL,
  max_samples = 10^5,
  size = 10^3,
  mcse_crit = 0.5
)

## S3 method for class 'marginals'
gibbs_ad_use(
  marginalized,
  finds,
  id = NULL,
  type = NULL,
  type_name = NULL,
  max_samples = 10^5,
  size = 10^3,
  mcse_crit = 0.5
)

Arguments

marginalized	A list object of class marginals, the output of gibbs_ad.
finds	Either the list object of finds used as input to produce marginals or a data.frame of two columns, the first listing the context and the second the incidence of the type in that context.
id	A vector of the id of one or more specific finds whose use date is to be estimated. The values of id must match those in the list of finds. If type is used, id is ignored.
type	A vector of one or more types to estimate a use density for. Must contain a value if id is NULL.
type_name	A customized label for the type (e.g., if one is selecting via id or has combined subtypes). If only type is used to select finds, the default will be that label Otherwise the default is simply "Type."
max_samples	Maximum number of samples to run. Default is 10^5.
size	The number of samples to take on each iteration of the main Gibbs sampler. Default is 10^3.
mcse_crit	Criterion for the Monte Carlo standard error to stop the Gibbs sampler. Only the MCSE of the use date is used as a stopping rule.

Details

Depending on whether one is using id numbers or type(s), the id or type argument is used, which takes a vector of the entries' names. The gibbs_ad_use function samples a use date between the production and depositional densities from the results of gibbs_ad, and in turn pools those densities for the production and deposition of the stipulated ids/type; the resulting list object does not express marginalized densities of production and deposition in light of the estimation of a use date.

See gibbs_ad for information on consistent batch means and Monte Carlo standard error, which are used to determined convergence for the use date.

Value

A list of class use_marginals of the density of a use date, conditional upon production and depositional dates.

Examples

x <- c("A", "B", "C", "D", "E", "F", "G", "H", "I", "J")
y <- c("B", "D", "G", "H", "K")
z <- c("F", "K", "L", "M")
contexts <- list(x, y, z)

f1 <- list(id = "find01", assoc = "D", type = c("type1", "form1"))
f2 <- list(id = "find02", assoc = "E", type = c("type1", "form2"))
f3 <- list(id = "find03", assoc = "G", type = c("type1", "form1"))
f4 <- list(id = "find04", assoc = "H", type = c("type2", "form1"))
f5 <- list(id = "find05", assoc = "I", type = "type2")
f6 <- list(id = "find06", assoc = "H", type = NULL)

artifacts <- list(f1, f2, f3, f4, f5, f6)
 
# external constraints
coin1 <- list(id = "coin1", assoc = "B", type = NULL, samples = runif(100,-320,-300))
coin2 <- list(id = "coin2", assoc = "G", type = NULL, samples = seq(37, 41, length = 100))
destr <- list(id = "destr", assoc = "J", type = NULL, samples = 79)

tpq_info <- list(coin1, coin2)
taq_info <- list(destr)

result <- gibbs_ad(contexts, finds = artifacts, tpq = tpq_info, taq = taq_info)

# use dates by specifying ids
gibbs_ad_use(result, artifacts, id = c("find04", "find05"), max_samples = 2000, mcse_crit = 2)

# use dates by specifying types
gibbs_ad_use(result, artifacts, type = "type1", max_samples = 2000, mcse_crit = 2)

eratosthenes documentation built on June 28, 2025, 1:08 a.m.