garchmodels (version 0.1.1)

garch_reg: General Interface for GARCH Models

Description

General Interface for GARCH Models

Usage

garch_reg(
  mode = "regression",
  arch_order = NULL,
  garch_order = NULL,
  ar_order = NULL,
  ma_order = NULL,
  tune_by = NULL
)

Arguments

mode

A single character string for the type of model.

arch_order

An integer giving the order of the ARCH part for the variance model.

garch_order

An integer giving the order of the GARCH part for the variance model.

ar_order

An integer giving the order of the AR part for the mean model.

ma_order

An integer giving the order of the MA part for the mean model.

tune_by

Default is set to NULL, when no tuning. If you want to tune, you must choose between "seriesFor" or "sigmaFor" options. This will cause the function to not return a nested tibble and be able to tune.

These arguments are converted to their specific names at the time that the model is fit.

Other options and argument can be set using set_engine() (See Engine Details below).

Value

A model specification

Engine Details

rugarch (default)

The engine uses rugarch::ugarchspec() and rugarch::ugarchfit().

Function Parameters:

## Formal class 'standardGeneric' [package "methods"] with 8 slots
##   ..@ .Data     :function (variance.model = list(model = "sGARCH", garchOrder = c(1, 1), 
##     submodel = NULL, external.regressors = NULL, variance.targeting = FALSE), 
##     mean.model = list(armaOrder = c(1, 1), include.mean = TRUE, archm = FALSE, 
##         archpow = 1, arfima = FALSE, external.regressors = NULL, archex = FALSE), 
##     distribution.model = "norm", start.pars = list(), fixed.pars = list(), 
##     ...)  
##   ..@ generic   : chr "ugarchspec"
##   .. ..- attr(*, "package")= chr "rugarch"
##   ..@ package   : chr "rugarch"
##   ..@ group     : list()
##   ..@ valueClass: chr(0) 
##   ..@ signature : chr [1:5] "variance.model" "mean.model" "distribution.model" "start.pars" ...
##   ..@ default   :Formal class 'derivedDefaultMethod' [package "methods"] with 4 slots
##   .. .. ..@ .Data  :function (variance.model = list(model = "sGARCH", garchOrder = c(1, 1), 
##     submodel = NULL, external.regressors = NULL, variance.targeting = FALSE), 
##     mean.model = list(armaOrder = c(1, 1), include.mean = TRUE, archm = FALSE, 
##         archpow = 1, arfima = FALSE, external.regressors = NULL, archex = FALSE), 
##     distribution.model = "norm", start.pars = list(), fixed.pars = list(), 
##     ...)  
##   .. .. ..@ target :Formal class 'signature' [package "methods"] with 3 slots
##   .. .. .. .. ..@ .Data  : chr "ANY"
##   .. .. .. .. ..@ names  : chr "variance.model"
##   .. .. .. .. ..@ package: chr "methods"
##   .. .. ..@ defined:Formal class 'signature' [package "methods"] with 3 slots
##   .. .. .. .. ..@ .Data  : chr "ANY"
##   .. .. .. .. ..@ names  : chr "variance.model"
##   .. .. .. .. ..@ package: chr "methods"
##   .. .. ..@ generic: chr "ugarchspec"
##   .. .. .. ..- attr(*, "package")= chr "rugarch"
##   ..@ skeleton  : language (new("derivedDefaultMethod", .Data = function (variance.model = list(model = "sGARCH",      garchOrder = c(1, 1),| __truncated__ ...

The Garch order for the variance model is provided using arch_order and garch_order parameters.. The ARMA order for the mean model is provided using ar_order and ma_order parameters. Other options and arguments can be set using set_engine().

#' Parameter Notes:

• xreg - This engine supports xregs for both the variance model and the mean model. You can do this in two ways, either enter the matrices through set_engine parameters or as a formula in fit (note that the latter option is more limited, since you will not be able to pass two different xregs, one for each model). For simpler cases this is a compact option.

• order parameters - The parameters of rugarch::ugarchspec are lists containing several elements, some of them the commands that are the main arguments of the function. If you want to modify the parameter that encompasses such a list, you must know that the parameter passed in the function parameter will always prevail. (See Examples).

Fit Details

Date and Date-Time Variable

It's a requirement to have a date or date-time variable as a predictor. The fit() interface accepts date and date-time features and handles them internally.

• fit(y ~ date)

Univariate (No xregs, Exogenous Regressors):

For univariate analysis, you must include a date or date-time feature. Simply use:

• Formula Interface: fit(y ~ date) will ignore xreg's.

Multivariate (xregs, Exogenous Regressors)

The xreg parameter is populated using the fit() function:

• Only factor, ordered factor, and numeric data will be used as xregs.

• Date and Date-time variables are not used as xregs

• character data should be converted to factor.

Xreg Example: Suppose you have 3 features:

1. y (target)

2. date (time stamp),

3. month.lbl (labeled month as a ordered factor).

The month.lbl is an exogenous regressor that can be passed to the garch_reg() using fit():

• fit(y ~ date + month.lbl) will pass month.lbl on as an exogenous regressor.

Note that date or date-time class values are excluded from xreg.

Details

Available engines:

• rugarch: Connects to rugarch::ugarchspec() first and then to rugarch::ugarchfit().

See Also

fit.model_spec(), set_engine()

Examples

# NOT RUN {
library(tidymodels)
library(garchmodels)
library(modeltime)
library(tidyverse)
library(timetk)
library(lubridate)

rIBM_extended <- rIBM %>%
    future_frame(.length_out = 24, .bind_data = TRUE) 

rIBM_train  <- rIBM_extended %>% drop_na()
rIBM_future <- rIBM_extended %>% filter(is.na(daily_returns))

model_garch_fit <-garchmodels::garch_reg(mode = "regression",
                                          arch_order = 1,
                                          garch_order = 1) %>%
    set_engine("rugarch") %>%
    fit(daily_returns ~ date, data = rIBM_train)

predict(model_garch_fit, rIBM_future)

model_garch_fit <-garchmodels::garch_reg(mode = "regression",
                                        arch_order = 2,
                                        garch_order = 2) %>%
    set_engine("rugarch", variance.model = list(model='gjrGARCH', 
                                                garchOrder=c(1,1)),
               mean.model     = list(armaOrder=c(0,0))) %>%
    fit(daily_returns ~ date, data = rIBM_train)

predict(model_garch_fit, rIBM_future)
# }