Implementation of rugarch methods for a GARCH(p,q) as R6 class

Implementation of rugarch methods for a GARCH(p,q) as R6 class

Implementation of rugarch methods for a GARCH(p,q) as R6 class

Note

Version 1.0.3

Public fields

control

List to contain custom control parameters.

model_name

Character, model name

Active bindings

archOrder

Numeric scalar, ARCH order.

garchOrder

Numeric scalar, GARCH order.

order

Numeric named vector, orders of the GARCH model. The first element is the ARCH order, while the second the GARCH order.

omega

Numeric scalar, intercept parameter.

alpha

Numeric vector, ARCH parameters.

beta

Numeric vector, GARCH parameters.

coefficients

Numeric vector, model's coefficients.

A

Numeric matrix, companion matrix.

b

Numeric vector.

d

Numeric vector

std.errors

Numeric vector, standard errors of the model's coefficients.

sigma2_inf

Numeric scalar, long-term unconditional std. deviation.

loglik

model log-likelihood

tidy

Tibble with estimated parameters and relative std. errors.

Methods

Public methods

• sGARCH$new()

• sGARCH$fit()

• sGARCH$filter()

• sGARCH$update()

• sGARCH$update_std.errors()

• sGARCH$next_step()

• sGARCH$print()

• sGARCH$clone()

---

Method new()

Initialize a standard GARCH model

Usage

sGARCH$new(archOrder = 1, garchOrder = 1, mode = "unitOmega")

Arguments

archOrder

Integer scalar, ARCH order.

garchOrder

Integer scalar, GARCH order.

mode

Character, one of "unitOmega", "targetSigma2", "freeOmega".

---

Method fit()

Fit the GARCH model with rugarch function.

Usage

sGARCH$fit(x, weights)

Arguments

x

Numeric, vector. Time series to be fitted.

weights

Numeric, vector. Optional custom weights.

---

Method filter()

Filter method from rugarch package to compute GARCH variance, residuals and log-likelihoods.

Usage

sGARCH$filter(x, eps0 = NULL, sigma20 = NULL)

Arguments

x

Numeric, vector. Time series to be filtered.

eps0

Optional numeric initial epsilons to prepend (length p+q).

sigma20

Optional numeric initial variances to prepend (length p+q).

---

Method update()

Update the coefficients of the model

Usage

sGARCH$update(coefficients)

Arguments

coefficients

Numeric, named vector. Model's coefficients.

---

Method update_std.errors()

Numerical computation of the std. errors of the parameters.

Usage

sGARCH$update_std.errors(std.errors)

Arguments

std.errors

Numeric std. errors.

---

Method next_step()

Next step GARCH std. deviation forecast

Usage

sGARCH$next_step(eps0, sigma20)

Arguments

eps0

Numeric initial epsilons to prepend (length p+q).

sigma20

Numeric initial variances to prepend (length p+q).

---

Method print()

Print method for GARCH_modelR6 class.

Usage

sGARCH$print()

---

Method clone()

The objects of this class are cloneable with this method.

Usage

sGARCH$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.