Are combination forecasts of S&P 500 volatility statistically superior?



ARFIMA(1,d,O) case reveals that volatility exhibits long memory properties.

In order to generate model based volatility forecasts which capture the in-
formation available at time
t efficiently, the volatility models were reestimated
for time-step
t using data from t 999 to t. The resulting parameter values
were then used to generate 22 day-ahead volatility forecasts (
t + 1 t + 22),
this time horizon is used for comparability with the
VIX IV forecast. The first
forecast period covers the trading period from 13 December 1993 to 12 January
1994. For subsequent forecasts the model parameters were reestimated using a
sliding estimation window of 1OOO observations. The last forecast period covers
18 September 2OO3 to 17 October 2OO3, leaving 246O forecasts.

3.2 Combining forecasts

Two strategies have been employed to construct combination forecasts. The
simplest, and most naïve approach sets the combinations to be the mean of the
constituent forecasts, thus an equally weighted combination of each forecast.

The alternative is to utilise the regression combination approach discussed
in Clements and Hendry (1998) where the combination weights are derived from
the following regression,

rvt+22 = ɑo + ɑi ft + s2 ft + ∙∙∙ + an ft + et            (7)

where RVt22 is the target volatility, the average RV over the 22 day forecast
horizon (
t +1 to t + 22) and ftz, г = 1, 2, ..., n are n different forecasts of average
volatility (
t + 1 to t + 22) formed at time t to be included in the combination.
The resulting combination forecast is then given by

ft = So + Si ft1 + ¾ ft2 + ∙∙∙ + an ftra                  (8)

1O



More intriguing information

1. Robust Econometrics
2. 5th and 8th grade pupils’ and teachers’ perceptions of the relationships between teaching methods, classroom ethos, and positive affective attitudes towards learning mathematics in Japan
3. Testing for One-Factor Models versus Stochastic Volatility Models
4. Handling the measurement error problem by means of panel data: Moment methods applied on firm data
5. What should educational research do, and how should it do it? A response to “Will a clinical approach make educational research more relevant to practice” by Jacquelien Bulterman-Bos
6. Placenta ingestion by rats enhances y- and n-opioid antinociception, but suppresses A-opioid antinociception
7. The name is absent
8. Achieving the MDGs – A Note
9. Popular Conceptions of Nationhood in Old and New European
10. Word searches: on the use of verbal and non-verbal resources during classroom talk
11. ISO 9000 -- A MARKETING TOOL FOR U.S. AGRIBUSINESS
12. Research Design, as Independent of Methods
13. The name is absent
14. The Nobel Memorial Prize for Robert F. Engle
15. How we might be able to understand the brain
16. Declining Discount Rates: Evidence from the UK
17. The name is absent
18. Target Acquisition in Multiscale Electronic Worlds
19. The Shepherd Sinfonia
20. The name is absent