期刊名称:Discussion Papers / Department of Economics, University of Essex
出版年度:2006
卷号:2006
出版社:University of Essex
摘要:When studying a time series of implied Risk Neutral Densities (RNDs) or other
implied statistics, one is faced with the problem of maturity dependence, given
that option contracts have a fixed expiry date. Therefore, estimates from
consecutive days are not directly comparable. Further, we can only obtain
implied RNDs for a limited set of expiration dates. In this paper we introduce
two new methods to overcome the time to maturity problem. First, we propose an
alternative method for calculating constant time horizon Economic Value at Risk
(EVaR), which is much simpler than the method currently being used at the Bank
of England. Our method is based on an empirical scaling law for the quantiles in
a log-log plot, and thus, we are able to interpolate and extrapolate the EVaR
for any time horizon. The second method is based on an RND surface constructed
across strikes and maturities, which enables us to obtain RNDs for any time
horizon. Removing the maturity dependence of implied RNDs and related statistics
is useful in many applications, such as in (i) the construction of implied
volatility indices like the VIX, (ii) the assessment of market uncertainty by
central banks (iii) time series analysis of EVaR, or (iv) event studies.