Modelling and Forecasting Volatility with VIX index

Modelling and forecasting volatility with VIX index
The main objective of our study is to model and forecast volatility as measured by the VIX index, with
the aim of producing information to banks and also to macroeconomists.
The VIX index introduced in 1993, derived from quote options on SP500 index is largely followed by
financial and economic deciders. It reflects markets sentiment and risk aversion. Since volatility is
said to have increased in the past months, due mainly to the Euro crisis, it is interesting and relevant
to check and explain this possible outcome. Apart from the necessary econometric analysis of the
index to test for linearity and stationnarity, we used, in our model building phase, several time
periods in order to determine the robustness of the results and stability of coefficients over time. We
suspect a break in the volatility pattern between August 2007 till March 2009 and again after April
We begin with Arma/Arima models, augmented with Garch errors and exogenous regressors : some
macroeconomic leading indicators were indeed included in regressions. In particular we tested the
effects of negative returns since they are likely to increase volatility. We also tested the ECB deposit
facilities volumes which account for the liquidity shortages and confidence losses in the market, as
well as some announcements regarding the debt refinancing in the European zone. We conclude
with a possible forecasting model.