MSfit: Markov-switching copula model.

In LucasDowiak/dynocopula: Implementation of markov switching, smooth transition, and sequential break point copulas

Markov-switching copula model.

Description

MSfit estimates a markov switching copula on a bivariate time series of uniform marginal distributions.

Usage

MSfit(x, y, family = list(1, 1), initValues, tol = 1e-05)

Arguments

x	A numeric vector of uniform marginal values.
y	A numeric vector of uniform marginal values.
family	A list of integers specifying the family of the copula to use in each regime.
initValues	Sometimes optional numeric vector of starting values. See Details.
tol	A numeric value specifying the convergence tolerance of the model. Specifically, the model reaches convergene if the difference in likelihood from successive models falls below tol. Defaults to 1e-5

Details

For initValues, if the same copula family is used for each regime, no initial values need to be supplied. If the user wants different copula families estimated in different regimes, initValues need to be supplied. For a model with K regimes, the order of the values should be provided as follows:

1. Copula parameters in the order they appear in family

2. K * (K - 1) transition variables: p_{1,1},...,p_{1,k-1},p_{2,1},...,p_{2,k-1},...,p_{k,k-1}

3. K - 1 initial state parameters: p_{0,1},...,p_{0,k-1}

Value

MSfit returns an S3 object of class markovCopula.

The summary, plot, coef, and logLik functions will, repectively, print a summarization of the output, a plot of dependence measures, extract model parameters, and extract the log-likelihood values.

An object of class markovCopula has the following components:

log.likelihood	log-likelihood value for the regime
pars	a vector of coefficients for the copula
N	the length of the time-series
solver	the final output from optim
regime.inference	the model's condition density, conditional probability, conditional forecasts, and the smoothed probabilities
copula	details of the estimated copulas in each regime
transition	the transition matrix and initial regime vector
nregimes	the number of regimes in the model

LucasDowiak/dynocopula documentation built on April 12, 2024, 1:32 p.m.