This job listing has expired and may no longer be relevant!
12 Feb 2018

Part-Time Garch(1,1) forecasting – fast job – Quick money

Wilhelm S – Posted by WilhelmS Anywhere

Job Description

I need some help with this GARCH(1,1) forecast. I want to forecast volatility, out of the sample. inputdata is a daily return series for an index.

My main question is, what is my code doing at the moment?

Is  it estimating volatility inside the sample or outside the sample?

When I set n.ahead= 22, does it forecast 22 days ahead the volatility and then jump into the next one.

If possible I would like to have a code, where I first determine a timeperiod for fitting the garch and then I want to forecast the monthly volatility for each month. So the model would do a rolling forecast by increasing the fit for garch and then forecast one month at a time.

Is the output “Garchsiz<[email protected]$modeldata$sigma” at the moment daily volatility or variance? How to convert it into a 22 days variance?

 

model<-ugarchspec(variance.model = list(model = “sGARCH”, garchOrder = c(1, 1)), mean.model = list(armaOrder = c(0, 0), include.mean = FALSE), distribution.model = “norm”)
modelfit<-ugarchfit(spec=model,data=inputdata)
data = siz[1:k, ,drop=FALSE]
spec = getspec(modelfit)

setfixed(spec) <- as.list(coef(modelfit))
forecast = ugarchforecast(spec, n.ahead = 22, n.roll = o, data = inputdata[1:k, ,drop=FALSE], out.sample = o)

sigma(forecast)
plot(forecast)

Garchsiz<[email protected]$modeldata$sigma
Garch<-data.frame(siz, Garchsiz)
write.csv(Garch, file = “Garchsiz.csv”)

 

How to Apply

Mail me

Job Categories: Data Scientist / Statistician. Job Types: Part-Time. Salaries: Less than $100,000.

Apply for this Job