Volatility co-movements: a time scale decomposition analysis

Cipollini, A; Muzzioli, S

Risultato della ricerca: Paper

Abstract

In this paper we are interested in detecting contagion from US to European stock market volatilities in the period immediately after the Lehman Brothers’ collapse. The analysis, based on a factor decomposition of the covariance matrix of implied and realized volatilities, is carried for different sub-samples (identified as normal and crisis periods) and across different (high) frequency bands. In particular, the analysis is split in two stages. In the first stage, we retrieve the time series of wavelet coefficients for each volatility series for high frequency scales, using the Maximal Overlapping Discrete Wavelet transform and, in a second stage, we apply Maximum Likelihood for a factor decomposition of the short-run covariance matrix.Given our focus on the standardized factor loadings associated with the US shock (to control for an heteroscedasticity bias), the empirical findings show no evidence of contagion from the US stock market volatility in realized volatility to all the European countries, while for implied volatility we find a weak evidence of contagion which depends on the scale (the only country not influenced being Netherlands).
Lingua originaleEnglish
Stato di pubblicazionePublished - 2014

Fingerprint

Time scales
Decomposition analysis
Comovement
Implied volatility
Covariance matrix
Contagion
Decomposition
Short-run
Factors
The Netherlands
Wavelet transform
Maximum likelihood
European stock markets
Heteroscedasticity
European countries
Stock market volatility
Coefficients
Realized volatility

Cita questo

Volatility co-movements: a time scale decomposition analysis. / Cipollini, A; Muzzioli, S.

2014.

Risultato della ricerca: Paper

@conference{ad5ad4ed7b4c475eba7d79a192b613d0,
title = "Volatility co-movements: a time scale decomposition analysis",
abstract = "In this paper we are interested in detecting contagion from US to European stock market volatilities in the period immediately after the Lehman Brothers’ collapse. The analysis, based on a factor decomposition of the covariance matrix of implied and realized volatilities, is carried for different sub-samples (identified as normal and crisis periods) and across different (high) frequency bands. In particular, the analysis is split in two stages. In the first stage, we retrieve the time series of wavelet coefficients for each volatility series for high frequency scales, using the Maximal Overlapping Discrete Wavelet transform and, in a second stage, we apply Maximum Likelihood for a factor decomposition of the short-run covariance matrix.Given our focus on the standardized factor loadings associated with the US shock (to control for an heteroscedasticity bias), the empirical findings show no evidence of contagion from the US stock market volatility in realized volatility to all the European countries, while for implied volatility we find a weak evidence of contagion which depends on the scale (the only country not influenced being Netherlands).",
author = "{Cipollini, A; Muzzioli, S} and {Lo Cascio}, Iolanda",
year = "2014",
language = "English",

}

TY - CONF

T1 - Volatility co-movements: a time scale decomposition analysis

AU - Cipollini, A; Muzzioli, S

AU - Lo Cascio, Iolanda

PY - 2014

Y1 - 2014

N2 - In this paper we are interested in detecting contagion from US to European stock market volatilities in the period immediately after the Lehman Brothers’ collapse. The analysis, based on a factor decomposition of the covariance matrix of implied and realized volatilities, is carried for different sub-samples (identified as normal and crisis periods) and across different (high) frequency bands. In particular, the analysis is split in two stages. In the first stage, we retrieve the time series of wavelet coefficients for each volatility series for high frequency scales, using the Maximal Overlapping Discrete Wavelet transform and, in a second stage, we apply Maximum Likelihood for a factor decomposition of the short-run covariance matrix.Given our focus on the standardized factor loadings associated with the US shock (to control for an heteroscedasticity bias), the empirical findings show no evidence of contagion from the US stock market volatility in realized volatility to all the European countries, while for implied volatility we find a weak evidence of contagion which depends on the scale (the only country not influenced being Netherlands).

AB - In this paper we are interested in detecting contagion from US to European stock market volatilities in the period immediately after the Lehman Brothers’ collapse. The analysis, based on a factor decomposition of the covariance matrix of implied and realized volatilities, is carried for different sub-samples (identified as normal and crisis periods) and across different (high) frequency bands. In particular, the analysis is split in two stages. In the first stage, we retrieve the time series of wavelet coefficients for each volatility series for high frequency scales, using the Maximal Overlapping Discrete Wavelet transform and, in a second stage, we apply Maximum Likelihood for a factor decomposition of the short-run covariance matrix.Given our focus on the standardized factor loadings associated with the US shock (to control for an heteroscedasticity bias), the empirical findings show no evidence of contagion from the US stock market volatility in realized volatility to all the European countries, while for implied volatility we find a weak evidence of contagion which depends on the scale (the only country not influenced being Netherlands).

UR - http://hdl.handle.net/10447/100347

M3 - Paper

ER -