tabula (version 1.2.0)

dateEvent: Date archaeological assemblages

Description

Experimental (see note).

Usage

dateEvent(object, ...)
# S4 method for CountMatrix
dateEvent(object, dates, level = 0.95,
  cutoff = 90, jackknife = FALSE, bootstrap = FALSE, n = 1000, ...)

Arguments

object

A \(m \times p\) matrix of count data.

...

Further arguments to be passed to CA.

dates

A list of numeric dates. Dates will be matched with assemblage by names.

level

A length-one numeric vector giving the confidence level.

cutoff

An integer giving the cumulative percentage of variance used to select CA factorial components for linear model fitting (see details). All compounds with a cumulative percentage of variance of less than the cutoff value will be retained.

jackknife

A logical scalar: should the model be checked by jackknife estimation (removing each fabric/type one at a time)?

bootstrap

A logical scalar: should the model be checked by bootstrap resampling?

n

A non-negative integer giving the number of partial bootstrap replications (see details).

Value

An object of class '>DateModel.

Model checking

Resampling methods can be used to check the stability of the resulting model. If jackknife is TRUE, one type/fabric is removed at a time and all statistics are recalculated. In this way, one can assess whether certain type/fabric has a substantial influence on the date estimate. If bootstrap is TRUE, a large number of new bootstrap assemblages is created, with the same sample size, by resampling each of the original assemblage with replacement. Then, examination of the bootstrap statistics makes it possible to pinpoint assemblages that require further investigation.

The use of resampling options (jackknife and bootstrap) can lead to much longer execution times and larger output objects. To monitor the execution of these re-sampling procedures, a progress bar will automatically be displayed if the pbapply package is installed on your machine.

Details

This is an implementation of the chronological modeling method developed by Bellanger and Husi (2012, 2013).

This method allows the estimation of two probability densities. The first one (event date) represents the terminus post-quem of an archaeological assemblage: an event dated in calendar time. The second represents the "chronological profile" of the assemblage: the accumulation rate (Bellanger and Husi 2012).

This method - somewhat similar to that described by Poblome and Groenen 2003 - is based on the adjustment of a Gaussian multiple linear regression model on the factors resulting from a correspondence analysis. This model results from the known dates of a selection of reliable contexts and allows to predict the event dates of the remaining assemblage with a 95 confidence interval.

Since correspondence analysis allows the rows and columns of a contingency table to be projected in the same space (through the transition formula), it is possible to estimate the date of each fabric using the previous model. Finally, the accumulation date of each context is defined as the mean of the fabric dates, weighted by their relative proportions in that context (akin to the Mean Ceramic Date proposed by South 1977).

This method relies on strong archaeological and statistical assumptions. Use it if you know what you are doing (see references below).

References

Bellanger, L. & Husi, P. (2013). Mesurer et mod<U+00E9>liser le temps inscrit dans la mati<U+00E8>re <U+00E0> partir d'une source mat<U+00E9>rielle : la c<U+00E9>ramique m<U+00E9>di<U+00E9>vale. In Mesure et Histoire M<U+00E9>di<U+00E9>vale. Histoire ancienne et m<U+00E9>di<U+00E9>vale. Paris: Publication de la Sorbonne, p. 119-134.

Bellanger, L. & Husi, P. (2012). Statistical Tool for Dating and Interpreting Archaeological Contexts Using Pottery. Journal of Archaeological Science, 39(4), 777-790. DOI: 10.1016/j.jas.2011.06.031.

Bellanger, L., Tomassone, R. & Husi, P. (2008). A Statistical Approach for Dating Archaeological Contexts. Journal of Data Science, 6, 135-154.

Bellanger, L., Husi, P. & Tomassone, R. (2006). Une approche statistique pour la datation de contextes arch<U+00E9>ologiques. Revue de Statistique Appliqu<U+00E9>e, 54(2), 65-81.

Bellanger, L., Husi, P. & Tomassone, R. (2006). Statistical Aspects of Pottery Quantification for the Dating of Some Archaeological Contexts. Archaeometry, 48(1), 169-183. DOI: 10.1111/j.1475-4754.2006.00249.x.

Poblome, J. & Groenen, P. J. F. (2003). Constrained Correspondence Analysis for Seriation of Sagalassos Tablewares. In Doerr, M. & Apostolis, S. (eds.), The Digital Heritage of Archaeology. Athens: Hellenic Ministry of Culture.

South, S. A. (1977). Method and Theory in Historical Archaeology. New York: Academic Press.

See Also

refine

Examples

# NOT RUN {
# Coerce dataset to abundance (count) matrix
zuni <- as(zuni, "CountMatrix")

# Assume that some assemblages are reliably dated (this is NOT a real example).
# The names of the vector entries must match the names of the assemblages.
dates <- list(
  LZ0569 = 1097, LZ0279 = 1119, CS16 = 1328, LZ0066 = 1111,
  LZ0852 = 1216, LZ1209 = 1251, CS144 = 1262, LZ0563 = 1206,
  LZ0329 = 1076, LZ0005Q = 859, LZ0322 = 1109, LZ0067 = 863,
  LZ0578 = 1180, LZ0227 = 1104, LZ0610 = 1074
)

# Model the event and accumulation date for each assemblage.
(model <- dateEvent(zuni, dates, cutoff = 90))

# Plot results for the first five assemblages
# Plot event date distribution
plotDate(model, type = "event", select = 1:5) +
  ggplot2::theme_bw()
# Plot accumulation date distribution
plotDate(model, type = "acc", select = 1:5) +
  ggplot2::theme_bw()
# Plot both distributions
plotDate(model, select = "LZ1105") +
  ggplot2::theme_bw()

# Check model variability
checked <- dateEvent(zuni, dates, cutoff = 90,
                     jackknife = TRUE, bootstrap = TRUE, n = 1000)

# Extract results for the first 5 assemblages
## Modeled event dates
checked["rows", 1:5]
## Jackknife fabrics
checked["jackknife", 1:5]
## Bootstrap of assemblages
checked["bootstrap", 1:5]
# }