Neglecting parameter changes in GARCH models

Hillebrand, Eric

doi:10.1016/j.jeconom.2004.09.005

Cited by 280 publications

(144 citation statements)

References 37 publications

Supporting

Mentioning

132

Contrasting

Unclassified

Order By: Relevance

“…Recently, it has been suggested that either long memory (Mikosch and Starica, 2004) or parameter changes (Hillebrand, 2005) in the data generating process can give the impression of IGARCH model. …”

Section: The Integrated Garch Modelsmentioning

confidence: 99%

Volatility persistence in crude oil markets

Charles

Darné

2014

Energy Policy

View full text Add to dashboard Cite

show abstract

Section: The Integrated Garch Modelsmentioning

confidence: 99%

Volatility persistence in crude oil markets

Charles

Darné

2014

Energy Policy

View full text Add to dashboard Cite

show abstract

“…Failure to account for structural breaks in the unconditional variance of stock market returns can lead to sizeable upward biases in the degree of persistence in estimated GARCH models Hillebrand, 2005; building on earlier work by Diebold, 1986;Hendry, 1986;. With structural breaks GARCH models do not accurately track changes in the unconditional variance leading to forecasts that underestimate or overestimate volatility on average for long stretches.…”

Section: Introductionmentioning

confidence: 99%

Structural breaks and GARCH models of stock return volatility: The case of South Africa

et al. 2012

View full text Add to dashboard Cite

This paper investigates the empirical relevance of structural breaks in forecasting stock return volatility using both in-sample and out-of-sample tests and daily returns for the Johannesburg Stock Exchange (JSE) All Share Index from 07/02/1995 to 08/25/2010. We find evidence of structural breaks in the unconditional variance of the stock returns series over the period, with high levels of persistence and variability in the parameter estimates of the GARCH (1, 1) model across the sub-samples defined by the structural breaks. This indicates that structural breaks are empirically relevant to stock return volatility in South Africa. In out-of-sample tests, we find that combining forecasts from different benchmark and competing models that accommodate structural breaks in volatility improves the accuracy of volatility forecasting. Furthermore, for shorter horizons, the MS-GARCH model better captures asymmetry in stock return volatility than the GJR-GARCH (1, 1) model, which better suited to longer horizons, but in general, the asymmetric models fail to outperform the GARCH (1,1) model. Keywords: stock return volatility, structural breaks, in-sample tests, out-of-sample tests, GARCH Models.JEL classification: C22, C53, G11, G12 ♦ We would like to thank Professor Mehmet Bacilar for many helpful comments. However, any errors are solely ours.

show abstract

“…More recently, Mikosch and Stărică (2004) prove that the IGARCH model makes sense when non-stationary data reflect changes in the unconditional variance. Hillebrand (2005) shows that in the presence of neglected parameter change-points, even a single deterministic change-point can cause GARCH to measure volatility persistence inappropriately. Alternatively, Hamilton and Susmel (1994) and Kim et al (1998) suggest that the long-run variance dynamics may include regime shifts, but within a given regime, it may follow a GARCH process.…”

Section: Introductionmentioning

confidence: 99%

“…First, on the one hand, Stock and Watson (2003), Bhar andHamori (2003), Mills andWang (2003), and Summers (2005) document the decline in volatility of Japan's output growth. On the other hand, Diebold (1986), Lamoureux and Lastrapes (1990), Mikosch andStărică (2004), andHillebrand (2005) note theoretically that Engle and Bollerslev's (1986) IGARCH may result from instability of the constant term in the conditional variance (i.e., nonstationarity of the unconditional variance). Neglecting such changes can generate spuriously measured persistence of the conditional variance heavily biased towards one.…”

mentioning

confidence: 99%

“…In addition, the high persistence measures may reflect structural changes in the mean or variance of growth rates, which the GARCH estimations ignore (Diebold, 1986;Lamoureux and Lastrapes, 1990;Mikosch and Stărică, 2004;Hillebrand, 2005;Krämer and Azamo, 2007;and Fang and Miller, 2008). Following Franses and Ghijsels (1999) and Charles and Darné (2005), we, first, employ the method of Chen and Liu (1993) to detect and correct for additive outliers (AOs) and 6 Balke and Fomby (1994) analyze fifteen post-World War II U.S. macroeconomic time series using the outlier identification procedure based on Tsay (1988) and find that outliers may prove important for U.S. macroeconomic data, and such aberrant observations may lead to large ARCH test statistics.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Modeling the volatility of real GDP growth: The case of Japan revisited

Fang

Miller

2009

Japan and the World Economy

View full text Add to dashboard Cite

Previous studies (e.g., Hamori, 2000;Ho and Tsui, 2003;Fountas et al., 2004) find high volatility persistence of economic growth rates using generalized autoregressive conditional heteroskedasticity (GARCH) specifications. This paper reexamines the Japanese case, using the same approach and showing that this finding of high volatility persistence reflects the Great Moderation, which features a sharp decline in the variance as well as two falls in the mean of the growth rates identified by Perrons (1998, 2003) multiple structural change test. Our empirical results provide new evidence. First, excess kurtosis drops substantially or disappears in the GARCH or exponential GARCH model that corrects for an additive outlier. Second, using the outlier-corrected data, the integrated GARCH effect or high volatility persistence remains in the specification once we introduce intercept-shift dummies into the mean equation. Third, the time-varying variance falls sharply, only when we incorporate the break in the variance equation. Fourth, the ARCH in mean model finds no effects of our more correct measure of output volatility on output growth or of output growth on its volatility. Journal of Economic Literature Classification: C32, E32, O40

show abstract

Neglecting parameter changes in GARCH models

Cited by 280 publications

References 37 publications

Volatility persistence in crude oil markets

Volatility persistence in crude oil markets

Structural breaks and GARCH models of stock return volatility: The case of South Africa

Modeling the volatility of real GDP growth: The case of Japan revisited

Contact Info

Product

Resources

About