Chapter 2. Univariate Volatility Modeling (in R/Julia)


Copyright 2011 - 2019 Jon Danielsson. This code is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This code is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. The GNU General Public License is available at: https://www.gnu.org/licenses/.

The original 2011 R code will not fully work on a recent R because there have been some changes to libraries. At least two R packages support estimating GARCH style models. rugarch by Alexios Ghalanos and fGarch. For our purposes there nothing to separate them but rugarch is regularly maintained, but fGarch appears not to be.


Listing 2.1/2.2: ARCH and GARCH estimation in R
Last updated August 2019

library(rugarch)
p = read.csv('index.csv')
y=diff(log(p$Index))*100
y=y-mean(y)
## We multiply returns by 100 and de-mean them
spec1 = ugarchspec(variance.model = list( garchOrder = c(1, 1)),
 mean.model = list( armaOrder = c(0,0),include.mean = FALSE))
res1 = ugarchfit(spec = spec1, data = y)
spec2 = ugarchspec(variance.model = list( garchOrder = c(1, 0)),
 mean.model = list( armaOrder = c(0,0),include.mean = FALSE))
res2 = ugarchfit(spec = spec2, data = y)
spec3 = ugarchspec(variance.model = list( garchOrder = c(1, 1)),
 mean.model = list( armaOrder = c(0,0),include.mean = FALSE),
 distribution.model = "std")
res3 = ugarchfit(spec = spec3, data = y)
## plot(res) shows various graphical analysis, works in command line
		
Listing 2.1/2.2: ARCH and GARCH estimation in Julia
Last updated June 2018

## No package for ARCH/GARCH estimation as of June 2018
## We have a mini-package for estimating GARCH(1,1) and t-GARCH(1,1)
using CSV;
p = CSV.read("index.csv", nullable = false)
y = diff(log.(p[:,1]))*100
y = y - mean(y)
using FRFGarch;
## Ensure that the folder FRFGarch is in the folder returned when LOAD_PATH is keyed into Julia command line
## The package must be in ../FRFGarch/src/FRFGarch.jl
res = GARCHfit(y, "GARCH");
println(res.par)
res_t = GARCHfit(y, "t-GARCH");
println(res_t.par)
## GARCH estimation will be slightly different from other languages
## this is due to GARCHfit choosing initial conditional vol = sample vol
		

Listing 2.3/2.4: Advanced ARCH and GARCH estimation in R
Last updated August 2019

library(rugarch)
## normal APARCH(1,1)
spec4 = ugarchspec(variance.model = list(model="apARCH", garchOrder = c(1, 1)),
 mean.model = list( armaOrder = c(0,0),include.mean = FALSE))
res4 = ugarchfit(spec = spec4, data = y)
#show(res4)
spec5 = ugarchspec(variance.model = list(model="apARCH", garchOrder = c(1, 1)),
 mean.model = list( armaOrder = c(0,0),include.mean = FALSE), fixed.pars=list(delta=2))
res5 = ugarchfit(spec = spec5, data = y)
show(res5)
		
Listing 2.3/2.4: Advanced ARCH and GARCH estimation in Julia
Last updated June 2018

## We use the same mini-package to estimate APARCH(1,1)
res_AP = GARCHfit(y, "APARCH");
println(res_AP.par)
## GARCH estimation will be slightly different from other languages
## this is due to GARCHfit choosing initial conditional vol = sample vol