Forecasting value at risk and expected shortfall with mixed data sampling

Le, Trung H.

doi:10.1016/j.ijforecast.2020.01.008

Cited by 32 publications

(31 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, they call for additional research because our study naturally cannot cover all potential alternative estimation models. There is a variety of additional techniques which do not require a full specification of the conditional distribution of the data and may thus reduce misestimation risk (rejections) below the levels observed for our best models (see Cotter & Dowd, 2010; Escanciano & Mayoral, 2009; Le, 2020). Unfortunately, before such alternatives can be tested within a framework of the Du and Escanciano (2017) type, it needs to be enhanced beyond a setup bound to invertible distribution functions.…”

Section: Discussionmentioning

confidence: 99%

“…When it comes to volatility modeling, the literature tells us that it is unlikely to find a model that perfectly describes our data (see Bollerslev, 1986) and that, out of hundreds of competing models, the simple GARCH(1, 1) model tends to perform best (see Hansen & Lunde, 2005). Therefore, supplemented by the fact that higher‐order autocorrelation of losses is typically negligible (see Campbell et al, 1993), researchers and practitioners often prefer AR(1)–GARCH(1, 1) settings (see Auer, 2015; Du & Escanciano, 2017; Le, 2020; McNeil & Frey, 2000). We follow this majority approach and use the specification

μ_{t} = α_{0} + α_{1} L_{t - 1},

σ_{t}^{2} = β_{0} + β_{1} {MathClass-open(σ_{t - 1} X_{t - 1} MathClass-close)}^{2} + β_{2} σ_{t - 1}^{2},

where

α_{0}, α_{1}

and

β_{0}, β_{1}, β_{2}

are the parameters of the mean and variance equation, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…In a recent article, Du and Escanciano (2017) make an important contribution to ending the backtesting debate for ES by developing unconditional and conditional coverage tests (with suitable size and power properties), which are easy to implement and to comprehend because they use ideas similar to the established VaR backtests. In many fields, these new tests are receiving significant attention because, after years of theoretical research, they finally allow a judgment of available ES estimators (see Hoga, 2019; Le, 2020; Novales & Garcia‐Jorcano, 2019). In the commodity context, there is now no more reason to put research on hold.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Time‐varying dynamics of expected shortfall in commodity futures markets

Mehlitz

Auer

2021

Journal of Futures Markets

View full text Add to dashboard Cite

Motivated by the growing interest of investors in commodities and by advances in risk measurement, we present a full‐scale analysis of expected shortfall (ES) in commodity futures markets. Besides illustrating the dynamics of historic ES, we evaluate whether popular estimators are suitable for forecasting future ES. By implementing a new backtest, we find that the performance of estimators hinges on market stability. Estimators tend to fail when markets are in turmoil and accurate forecasts are urgently needed. Even though a kernel method performs best on average, our results advise against the use of established estimators for risk (and margin) prediction.

show abstract

Section: Discussionmentioning

confidence: 99%

μ_{t} = α_{0} + α_{1} L_{t - 1},

σ_{t}^{2} = β_{0} + β_{1} {MathClass-open(σ_{t - 1} X_{t - 1} MathClass-close)}^{2} + β_{2} σ_{t - 1}^{2},

where

α_{0}, α_{1}

and

β_{0}, β_{1}, β_{2}

are the parameters of the mean and variance equation, respectively.…”

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Time‐varying dynamics of expected shortfall in commodity futures markets

Mehlitz

Auer

2021

Journal of Futures Markets

View full text Add to dashboard Cite

show abstract

“…Based on this new approach, the results provided a unified and consistent framework for introducing time-varying parameters in a wide class of non-linear models. Le (2020) develops a mixing data sampling (MIDAS) framework for forecasting VaR and ES. The methods of this author exploit the serial dependence on short-horizon returns, to directly forecast the tail dynamics of the desired horizon.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Since the number of recent contributions related to forecasting and backtesting ES and VaR is extremely large, the main contribution of this study is the estimation and forecasting of VaR and ES intervals through the use of SARIMA-GAS-GEVD. Nonetheless, Le (2020) used mixed data sampling (MIDAS) framework to forecast VaR and ES. The new methods of this author exploit the serial dependence on short-horizon returns to directly forecast the tail dynamics of the desired horizon.…”

Section: Introductionmentioning

confidence: 99%

Bootstrapping Time-Varying Uncertainty Intervals for Extreme Daily Return Periods

Makatjane

2022

IJFS

View full text Add to dashboard Cite

This study aims to overcome the problem of dimensionality, accurate estimation, and forecasting Value-at-Risk (VaR) and Expected Shortfall (ES) uncertainty intervals in high frequency data. A Bayesian bootstrapping and backtest density forecasts, which are based on a weighted threshold and quantile of a continuously ranked probability score, are developed. Developed backtesting procedures revealed that an estimated Seasonal autoregressive integrated moving average-generalized autoregressive score-generalized extreme value distribution (SARIMA–GAS–GEVD) with a skewed student-t distribution had the best prediction performance in forecasting and bootstrapping VaR and ES. Extension of this non-stationary distribution in literature is quite complicated since it requires specifications not only on how the usual Bayesian parameters change over time but also those with bulk distribution components. This implies that the combination of a stochastic econometric model with extreme value theory (EVT) procedures provides a robust basis necessary for the statistical backtesting and bootstrapping density predictions for VaR and ES.

show abstract

Forecasting value at risk and expected shortfall using high‐frequency data of domestic and international stock markets

Wang

Cheng

2022

Journal of Forecasting

View full text Add to dashboard Cite

This paper first proposes the generalized autoregressive score (GAS) factor X model for predicting Value‐at‐Risk (VaR) and expected shortfall (ES). This model is flexible for adding explanatory variables (X) that are informative for forecasting extreme risks and thus can improve forecast accuracy of the traditional GAS factor model. Then, to investigate whether high‐frequency data of domestic and international stock market are predictive of the extreme risk of Chinese stock market, X is set to be the corresponding realized volatility (RV) of these markets. Empirical studies based on 13 stock markets show that lagged RVs of domestic and influential international markets have prediction ability of extreme risks in Chinese stock market. The RV of the domestic market is more informative than that of international markets, among which the RVs of Asian and US markets are more predictive than European markets. We have also investigated compound information of the RVs of all markets, the results of which show that the mean of RVs is more informative than the first principle component; however, it is less informative than the individual RV of some markets. Additionally, there is no gain in forecast accuracy of the GAS factor RV models by forecast combination.

show abstract

Forecasting value at risk and expected shortfall with mixed data sampling

Cited by 32 publications

References 42 publications

Time‐varying dynamics of expected shortfall in commodity futures markets

Time‐varying dynamics of expected shortfall in commodity futures markets

Bootstrapping Time-Varying Uncertainty Intervals for Extreme Daily Return Periods

Forecasting value at risk and expected shortfall using high‐frequency data of domestic and international stock markets

Contact Info

Product

Resources

About