Excessive variation in risk-factor correlations and volatilities
β Scribed by Turan G. Bali; Hans Genberg; Salih N. Neftci
- Publisher
- John Wiley and Sons
- Year
- 2002
- Tongue
- English
- Weight
- 447 KB
- Volume
- 22
- Category
- Article
- ISSN
- 0270-7314
No coin nor oath required. For personal study only.
β¦ Synopsis
Abstract
This article explores the timeβseries behavior of correlations of returns, volatilities of returns,
volatilities of volatilities, and correlations of volatilities in domestic and international financial markets
such as equity (indices), interest rates (bonds), and currency (exchange rates)
using a Kalman filter approach to estimate the aforementioned parameters. The main findings include the
following. First, the correlations of risk factors are highly unstable over time, both in terms of sign and
absolute value. Second, the time variation of riskβfactor volatilities is stochastically nonlinear. Long
periods of deterministic volatilities are interrupted by sudden bursts of highly volatile periods. Third, daily
volatilities of riskβfactor volatilities fluctuate over time within a narrower band. Fourth, the
correlations between volatilities are generally positive and relatively stable over time. These results have
implications for financial risk management, dynamic asset allocation, and valuation of derivative securities.
Β© 2002 Wiley Periodicals, Inc. Jrl Fut Mark 22:1119β1146, 2002
π SIMILAR VOLUMES
## Background: A variety of factors are predictors of breast cancer risk. however, the studies conducted to establish these risk factors have rarely included african american women. the few studies with sufficient numbers of african-american women suggest that risk factors for breast cancer among a
## Abstract Using highβfrequency returns, realized volatility and correlation of the NYMEX light, sweet crude oil, and HenryβHub natural gas futures contracts are examined. The unconditional distributions of daily returns and daily realized variances are nonβGaussian, whereas the distributions of t