Abstract:Conditional VaR models′correct specification test is equivalent to testing the de-meaned hit sequence following a mar-tingale difference sequence (m.d.s), however the commonly used backtesting techniques only test some properties of the se-quence.Using generalized spectral test which directly tests the m.d.s property of the de-meaned hit sequence, we evaluate the out-of-sample predictive performance of various parametric, nonparametric and semi-parametric VaR models with a total of 22 models calculated by using rolling predictive method for China′s stock markets including Shanghai Composite Index, Hang Seng Index and Taiwan Weighted Index.Because conditional VaR models′correct specification test can not reflect the tail risk infor-mation exceeding one specific VaR level, in order to avoid the occurrence of extreme losses as well as to increase the robustness of the results, we adopt MCS ( model confidence set) test simultaneously by selecting the asymmetric loss functions proposed by Koenker and Bassett and the magnitude loss function proposed by Lopez.Comparing with SPA( Superior Predictive Ability) test, the main advantage of MCS test is that it does not require a benchmark model to be specified as is the case for SPA tests.It char-acterizes the entire set of models that are not significantly outperformed by other models, while a test for SPA only provides evi-dence about the relative performance of a single model ( the benchmark) . The empirical results imply the following three conclusions:①it would cause wrong results using the commonly applied backtest-ing techniques such as Kupiec likelihood ratio test, Christoffersen likelihood ratio test and Engle and Manganelli dynamic quantile test.However adopting generalized spectral test and MCS test with Lopez loss function simultaneously would give us more accu-rate and robust results.②Comparing with historical simulation models, extreme value theory models, CAViaR and CARE mod-els, the out-of-sample predictive performance of the GARCH family with student-t distribution is the best at 1%and 5%signifi-cant level during the financial crisis for these three stock indexes.This implies that the risk characteristics of mainland stock mar-ket is getting closer and closer to the mature stock markets of Hong Kong and Taiwan after more than 20 years development.③At 1%significant level, the optimal VaR predictive models of Hang Seng Index include one of the CARE models which can be used to measure extreme loss situation with small probability.This implies that price limit system implemented by Hong Kong yet not by mainland and Taiwan will make Hong Kong′s stock market face more risk during the financial crisis.

VaR,Expected Shortfall and Measurement of Consistency Risk

Conditional Tail-Related Risk Estimation Using Composite Asymmetric Least Squares and Empirical Likelihood

Expected Shortfall Or Median Shortfall

On the coherence of expected shortfall

On Exactitude in Financial Regulation: Value-at-Risk, Expected Shortfall, and Expectiles

Quantitative Models Improved on VaR to Measure Operational Risk

Estimating Value at Risk and Expected Shortfall: A Brief Review and Some New Developments

Risk measures induced by efficient insurance contracts

Risk Measures Based on Benchmark Loss Distributions

Evaluating Predictive Performance of Value-at-Risk Models Based on Generalized Spectrum and MCS Tests

Diversification quotients based on VaR and ES

Worst-Case Range Value-at-Risk with Partial Information

A Semi-parametric Realized Joint Value-at-Risk and Expected Shortfall Regression Framework

Interval-valued risk measure models and empirical analysis.

The problem of portfolio optimization based on safety-first criteria and the measure of VaR.

Mixed--frequency quantile regressions to forecast Value--at--Risk and Expected Shortfall

Value-at-Risk- and Expectile-based Systemic Risk Measures and Second-order Asymptotics: With Applications to Diversification

One Method of Application of Extreme Value Theory to Computing Value-at-Risk

ESTIMATION RISK IN GARCH VaR AND ES ESTIMATES

Risk Measures: Robustness, Elicitability, and Backtesting.

Estimation of CVaR and Empirical Analysis Based on Tail Index Regression Model