### Abstract

Lingua originale | English |
---|---|

Stato di pubblicazione | Published - 2013 |

### Fingerprint

### Cita questo

*Wavelet analysis of variance risk premium spillovers*.

**Wavelet analysis of variance risk premium spillovers.** / Cipollini, A; Muzzioli, S.

Risultato della ricerca: Paper

}

TY - CONF

T1 - Wavelet analysis of variance risk premium spillovers

AU - Cipollini, A; Muzzioli, S

AU - Lo Cascio, Iolanda

PY - 2013

Y1 - 2013

N2 - In this paper we construct a variance risk premium spillover index among France, Germany, UK, Switzerland and the US. The variance risk premium is measured by the difference between the difference between the (square) of implied volatility and expected realized variance of the stock market for next month. We also construct a spillover index for the constituents of the variance risk premium. The series under investigation exhibit long memory properties. The construction of a total spillover indicator suggested by Diebold-Yilmaz (2009) would then rely on modeling a fractionally integrated Vector Autoregressive Model, which might be subject to errors in specifying the correct lag length and the fractional differencing parameters. In order to avoid such misspecification errors, we employ wavelet analysis. In particular, we employ the Maximal Overlapping Transform and we compute the covariance matrix at different scales (associated to a frequency range). The spillover index is then obtained from the relative contribution of each (orthogonalized) shock to the variance of the other series at given scale, interpreted as a given investment time horizon.

AB - In this paper we construct a variance risk premium spillover index among France, Germany, UK, Switzerland and the US. The variance risk premium is measured by the difference between the difference between the (square) of implied volatility and expected realized variance of the stock market for next month. We also construct a spillover index for the constituents of the variance risk premium. The series under investigation exhibit long memory properties. The construction of a total spillover indicator suggested by Diebold-Yilmaz (2009) would then rely on modeling a fractionally integrated Vector Autoregressive Model, which might be subject to errors in specifying the correct lag length and the fractional differencing parameters. In order to avoid such misspecification errors, we employ wavelet analysis. In particular, we employ the Maximal Overlapping Transform and we compute the covariance matrix at different scales (associated to a frequency range). The spillover index is then obtained from the relative contribution of each (orthogonalized) shock to the variance of the other series at given scale, interpreted as a given investment time horizon.

UR - http://hdl.handle.net/10447/79976

UR - http://www.efmaefm.org/0EFMAMEETINGS/EFMA%20ANNUAL%20MEETINGS/2013-Reading/papers/EFMA2013_0440_fullpaper.pdf

M3 - Paper

ER -