Volatility Forecast Comparison Using Imperfect Volatility Proxies

By Andrew J. Patton. 2006

1,203 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/082b716b977dc218b416a8a2072fb755f8c0d24a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

On the Forecasting Accuracy of Multivariate GARCH Models (2010, background): “Building on Hansen and Lunde (2006a) and Patton (2009), Laurent, Rombouts, and Violante (2009) address these questions in the case of the comparison of multivariate volatility models using statistical loss functions. They show that the substitution of the underlying volatility by a proxy may induce ”
MGARCH models: Trade-off between feasibility and flexibility (2018, background): “…the estimation of realized covariances and correlations suffers from asynchronous trading and market microstructure noise, causing the covariance and correlation estimators to be biased and inconsistent; see, for instance, McAleer and Medeiros (2008), Patton (2011) and Corsi and Audrino (2012).”
Exogenous drivers of Bitcoin and Cryptocurrency volatility – A mixed data sampling approach to forecasting (2019, methodology): “(12) The two measures are often used to evaluate GARCH models (e.g. Bollerslev & Ghysels, 1996, Patton, 2011).”
Forecasting volatility and co-volatility of crude oil and gold futures: Effects of leverage, jumps, spillovers, and geopolitical risks (2020, methodology): “For comparing the out-of-sample forecasts, we extend the idea of Patton (2011) for univariate volatility models.”
Making and Evaluating Point Forecasts (2009, methodology): “table 1 of Patton (2011), which were obtained by Monte Carlo meth-”
Robust Forecasting of Dynamic Conditional Correlation GARCH Models (2013, methodology): “Loss function: In the presence of outliers (or jumps), Preminger and Franck (2007) recommend using forecast performance evaluation criteria that are less sensitive to extreme observations.”
ESTIMATING THE PERSISTENCE AND THE AUTOCORRELATION FUNCTION OF A TIME SERIES THAT IS MEASURED WITH ERROR (2013, unknown): “See also Patton (2011) and Patton and Sheppard (2009), who provide further insight about this issue.”
Decoupling the Short- and Long-Term Behavior of Stochastic Volatility (2016, methodology): “We observed that failing to forecast the volatility bursts during such episodes leads to disproportionately large losses with both MSE and QLIKE, and especially with MSE, these individual losses can represent a significant proportion of the total loss.”