Nell'ambito del ciclo di seminari del Dipartimento di Scienze per l'Economia e l'Impresa (DISEI) dell'Universita' di Firenze, il giorno
Martedì 14 Febbraio 2017 (Edificio D6/Aula Bracco, ossia 1.12):
si terranno i seguenti 2 seminari
ore 10 Francesca Lilla (Università di Bologna)
OPTION PRICING WITH HIGH FREQUENCY ESTIMATES OF CONTINUOUS AND DISCONTINUOUS VOLATILITY COMPONENTS
ore 11 Susana Martins, University of Minho & NIPE, Braga, Portugal
MODELLING SOVEREIGN DEBT CONTAGION: A SMOOTH TRANSITION APPROACH
Tutti gli interessati sono cordialmente invitati a partecipare. La lista dei seminari Disei è raggiungibile al seguente indirizzo: http://www.disei.unifi.it/vp-104-seminari.html
Abstract dei seminari
OPTION PRICING WITH HIGH FREQUENCY ESTIMATES OF CONTINUOUS AND DISCONTINUOUS VOLATILITY COMPONENTS
Jump-diffusion model represents the popular approach to match return stylized facts. Its estimation is simplified since prices are observable quantities, but it does not appear to be sufficient to yield the rapid increases in volatility which have been historically experienced. Examining price evolution at ultra high frequency, traditional measures of jump variation tend to spuriously assign a burst in volatility to the jump components. What jump-diffusion model identifies as genuine price jumps are period of heightened volatility. Starting from these findings, I develop a new multifactor volatility model, labeled as VARG-J, in which volatility is decomposed in two components: continuous and discontinuous. The first involves small changes while the second occasional big moves. The VARG-J model belongs to the class of affine discrete-time models based on high frequency volatility measures. The pricing approach performed in this paper allows to incorporate risk premia in order to compensate for two new sources of risk concerning both volatility components, beyond the risk related to shocks in log-returns. Combining high-frequency returns and option data, the empirical analysis illustrates that volatility jump component is a key ingredient for reproducing the observed volatility smile and for improving the pricing performances of S&P 500 Index options.
MODELLING SOVEREIGN DEBT CONTAGION: A SMOOTH TRANSITION APPROACH
ABSTRACT: In this paper, we study the timing and extent of the European sovereign debt contagion across nine Eurozone countries from 2005 until 2016. In this setting, financial contagion is measured as a significant increase in cross-market correlations which are assumed to follow a Smooth Transition Conditional Correlation (STCC-) GARCH model. The cross-market correlations change smoothly between two extreme states according to an observable transition variable. This variable is specified both as a function of time and lagged returns. When using time, a pre-crisis and two crises sub-periods are identified. Our results suggest contagion effects within the periphery countries, notably from Greece and Portugal, and between the periphery and core countries.