CTW {BCT}

Context Tree Weighting (CTW) algorithm

Description

Computes the prior predictive likelihood of the data given a specific alphabet. This function is used in for change-point point/segmentation problems

Usage

CTW(input_data, depth, desired_alphabet = NULL, beta = NULL)

Arguments

input_data

the sequence to be analysed. The sequence needs to be a "character" object. See the examples section of the BCT/kBCT functions on how to transform any dataset to a "character" object.

depth

maximum memory length.

desired_alphabet

set containing the symbols of the process. If not initialised, the default set contains all the unique symbols which appear in the sequence. This parameter is needed for the segmentation problem where short segments might not contain all the symbols in the alphabet.

beta

hyper-parameter of the model prior. Takes values between 0 and 1. If not initialised in the call function, the default value is 1-2^-m+1, where m is the size of the alphabet; for more information see Kontoyiannis et al. (2020).

Value

returns the natural logarithm of the prior predictive likelihood of the data.

See Also

BCT, kBCT

Examples

# For the gene_s dataset with a maximum depth of 10 (with dafault value of beta):
CTW(gene_s, 10)

# With the ["0", "1", "2", "3"] alphabet
CTW(gene_s, 10, "0123")

# For custom beta (e.g. 0.8):
CTW(gene_s, 10, ,0.8)