Bayesian Context Trees (BCT) algorithm
Finds the maximum a posteriori probability (MAP) tree model.
BCT(input_data, depth, beta = NULL)
input_data |
the sequence to be analysed. The sequence needs to be a "character" object. See the examples section on how to transform any dataset to a "character" object. |
depth |
maximum memory length. |
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). |
returns a list object which includes:
Contexts |
MAP model given as a list object containing the contexts of its leaves. |
Results |
a dataframe with the following columns: prior probability, log(prior probability), posterior probability, log(posterior probability), number of leaves, maximum depth, BIC score, AIC score and maximum log-likelihood. |
# Finding the MAP model with maximum depth <= 10
# for the SP500 dataset (with default value beta):
BCT(SP500, 10)
# For custom beta (e.g. 0.7):
BCT(SP500, 10, 0.7)
# The type of the input dataset is "character"
# If the dataset is contained within a vector:
q <- c(1, 0, 1, 0, 0, 1, 0, 0, 1, 1, 1, 0, 1, 0)
# Convert a vector to a "character" object:
s <- paste(q, collapse = "")
BCT(s, 2)
# Reading a file using the readChar function
# Example 1: The dataset is stored in a .txt file
# fileName <- '~/example_data.txt' # fileName stores the path to the dataset
# s<-readChar(fileName, file.info(fileName)$size)
# Make sure that s does not contain any "\n" at the end of the string
# To remove last entry:
# s<-gsub('.$', '', s)
# To remove any unwanted characters (e.g. "\n"):
# s<-gsub('\n', '', s)
# Example 2: The dataset is stored in a .csv file
# fileName <- '~/example_data.csv' # fileName stores the path to the dataset
# s<-readChar(fileName, file.info(fileName)$size)
# Depending on the running environment,
# s might contain unwanted characters such as: "\n" or "\r\n".
# Remove any unwanted characters (e.g. "\r\n"):
# s<-gsub('\r\n', '', s)
# Always make sure that s does not contain any unwanted charactersPlease choose more modern alternatives, such as Google Chrome or Mozilla Firefox.