Skip to content
QuantCalc Tools
Directory

Parametric Var Expected Shortfall

Acerbi-Szekely Expected Shortfall Backtesting: Non-Elicitability & Z2 Math

Gneiting proved standalone Expected Shortfall lacks a strictly consistent scoring function, yet Acerbi and Szekely’s unconditional Z2 statistic directly backtests joint VaR-ES tail severity without nuisance parameters.

By Elena Rostova, FRM, CQFReview not recorded
Acerbi-Szekely Expected Shortfall Backtesting: Non-Elicitability & Z2 Math — QuantCalc Tools 2026 quantitative & supervisory data card
QuantCalc Tools (Parametric VaR, Expected Shortfall & Fat-Tail Risk Calibration): Direct Expected Shortfall Z1/Z2 backtests and Fissler-Ziegel joint elicitability.Source: QuantCalc Tools Institutional Research DeskLicense: CC-BY-4.0

When Gneiting (2011) proved mathematically that Expected Shortfall is not standalone elicitable—meaning no strictly consistent loss function exists whose minimization uniquely recovers true $ES\alpha$—critics initially claimed Basel III FRTB 97.5% Expected Shortfall could never be backtested! Acerbi & Székely (2014) and Fissler & Ziegel (2016) overturned that misconception completely: **backtesting (hypothesis testing against $H0$) is distinct from elicitability (point forecast ranking), and furthermore the joint pair $(VaR\alpha, ES\alpha)$ IS strictly jointly elicitable!** Compute baseline normal VaR and ES pairs in our Value at Risk (VaR) calculator and review our Parametric VaR & Expected Shortfall calibration guide.

Exact Acerbi-Székely $Z1$ and $Z2$ Backtesting Equations#

Let $Xt$ denote portfolio PnL on day $t \in \{1, \dots, T\}$ (negative values are losses), while $VaR{t,\alpha} > 0$ and $ES{t,\alpha} > 0$ denote positive loss forecasts at tail probability $p = 1 - \alpha = 0.025$ (for 97.5% ES), with breach indicator $It = \mathbb{I}{\{Xt + VaR{t,\alpha} < 0\}}$. Under a correctly specified conditional return distribution $H0$:

  1. Conditional Tail Severity ($Z_1$): Tests solely whether realized losses on breach days average $ES{t,\alpha}$, conditioning on $NT = \sum{t=1}^T It > 0$:

$$Z1 = \frac{1}{NT}\sum{t=1}^T \frac{Xt \cdot It}{ES{t,\alpha}} + 1, \qquad \mathbb{E}{H0}[Z1 \mid NT > 0] = 0$$

  1. Unconditional Joint Tail Test ($Z_2$): Replaces random $N_T$ in the denominator with expected hits $T \cdot (1 - \alpha)$, simultaneously testing breach frequency AND breach magnitude in one unified statistic:

$$Z2 = \frac{1}{T(1 - \alpha)}\sum{t=1}^T \frac{Xt \cdot It}{ES{t,\alpha}} + 1, \qquad \mathbb{E}{H0}[Z2] = 0$$

Worked Non-Parametric Acerbi-Szekely Z2 Test Statistic Audit#

Formula or model from the articleNot independently verified
Worked Acerbi-Szekely Z1 & Z2 Backtest (T = 250 Days, 97.5% ES => p = 0.025, T*p = 6.25 Expected Hits)
  • Observed Breaches over 250 Days: N_T = 7 Breach Days (Passes Kupiec 97.5% window [2, 12]!)
  • Normalized Breach Ratios (-Xt / ES{t,0.975} on the 7 breach days, where X_t < 0):
  • Day 19: -$3.40M / $2.50M = 1.360 Day 142: -$4.20M / $2.60M = 1.615
  • Day 54: -$2.95M / $2.40M = 1.229 Day 188: -$3.85M / $2.50M = 1.540
  • Day 91: -$3.10M / $2.50M = 1.240 Day 219: -$4.55M / $2.60M = 1.750
  • Day 112:-$3.60M / $2.55M = 1.412
  • Sum of Signed Ratios Sum(Xt * It / ES_t) = -(1.360 + 1.229 + 1.240 + 1.412 + 1.615 + 1.540 + 1.750) = -10.146
  • 1) Conditional Severity Statistic Z_1:
  • Z_1 = (-10.146 / 7) + 1 = -1.4494 + 1 = -0.4494 (< -0.26 critical -> TAIL SEVERITY UNDERSTATED!)
  • 2) Unconditional Joint Statistic Z_2 (Divides by Expected Hits T*(1-alpha) = 6.25):
  • Z_2 = (-10.146 / 6.25) + 1 = -1.6234 + 1 = -0.6234 (<< -0.70 5% normal critical? Close to 1% reject!)
  • Note: Under Student-t(5) Monte Carlo null at T=250, 5% critical value for Z_2 is approx -0.72;
  • under Gaussian null at T=250, 5% left-tail critical value for Z_2 is -0.608 -> REJECTS GAUSSIAN ES!

Look at what happened on those 7 breach days: Kupiec's 97.5% VaR frequency test passed effortlessly ($NT = 7$ vs $6.25$ expected, well below the 12-hit cutoff in Kupiec POF likelihood-ratio zones)! Yet on every single breach day, realized losses exceeded $ES{t,0.975}$ by $23\%$ to $75\%$, driving $Z1 = -0.4494$ and $Z2 = -0.6234$ deep into left-tail rejection territory! Sign convention reminder: negative $Z1$ or $Z2$ signals risk underestimation; positive values near $+1.0$ signal conservative over-margining.

Acerbi-Szekely Z2 unconditional Expected Shortfall backtest and Fissler-Ziegel scoring pipeline
Acerbi-Szekely Z2 unconditional Expected Shortfall backtest and Fissler-Ziegel scoring pipelineVerified Exhibit

Fissler-Ziegel (2016) Jointly Elicitable Scoring Function ($FZ_0$)#

What if a quantitative research team wants to rank two competing forecasting models—say, Student-t Expected Shortfall versus EVT Peaks-Over-Threshold GPD—on out-of-sample data without knowing the "true" data-generating distribution? Fissler, Ziegel & Gneiting (2016) derived the 0-homogeneous strictly consistent joint loss function (writing $v < 0$ for negative PnL quantile and $e \le v < 0$ for negative PnL ES): $$S{FZ0}(v, e, x) = -\frac{1}{e}\left(e - v + \frac{(v - x)\,\mathbb{I}{\{x \le v\}}}{1 - \alpha}\right) + \ln(-e) - 1$$ Minimizing average $S{FZ0}$ across out-of-sample days is mathematically guaranteed (via Diebold-Mariano tests) to select the model closest to the true conditional $(VaR\alpha, ES_\alpha)$ pair!

Comparing Expected Shortfall Backtesting & Elicitability Frameworks#

ES Backtest MethodRequires Full Distribution $F_t$?Tests Frequency + Magnitude?Small-Sample Critical ValuesPrimary Institutional Strength
Basel MAR32 Dual-VaR ProxyNo (97.5% & 99% quantiles)Indirectly via 2 quantilesExact BinomialSimple statutory desk gate
McNeil-Frey Excess Residual BootstrapNo (Tail residuals only)Conditional Magnitude onlyNon-parametric bootstrapIsolates pure GPD tail shape error
Acerbi-Székely $Z_1$ TestYes (For null MC criticals)Conditional Magnitude onlyInsensitive to fat-tail null!Immune to VaR frequency error
Acerbi-Székely $Z_2$ TestYes (Simulated predictive $F_t$)Yes (Joint VaR + ES)Stable (~$-0.61$ to $-0.74$)High power against both errors
Fissler-Ziegel $S_{FZ0}$ Diebold-MarianoNo (Point $vt, et$ only!)Yes (Comparative ranking)Asymptotic HAC t-statRanks rival engines without $H_0$
Statistical rejection power of Kupiec 99% VaR, Dual MAR32 VaR, and Acerbi-Szekely Z2 against fat-tailed misspecification
Statistical rejection power of Kupiec 99% VaR, Dual MAR32 VaR, and Acerbi-Szekely Z2 against fat-tailed misspecificationVerified Exhibit

Ridge Backtesting & Berkowitz PIT Tail Censored Likelihood#

In addition to $Z2$, Acerbi & Székely (2017) introduced Ridge Backtesting based on Fissler-Ziegel's elicitability ridge, while Du & Escanciano (2017) generalized Christoffersen's autocorrelation test to cumulative Probability Integral Transforms (PIT) $ut = Ft(Xt)$ inside the 2.5% tail. Combine $Z_2$ validation with Christoffersen Markov independence testing, benchmark against the FRTB-ES-975 2026 Standard under UK PRA model validation guidance, and explore the Parametric VaR & Expected Shortfall Topic Hub.

Frequently asked questions

Does non-elicitability mean Expected Shortfall cannot be backtested?

No. Elicitability concerns whether a standalone point forecast minimizes an expected scoring rule for relative model ranking. Hypothesis testing (backtesting) evaluates whether realized tail losses match a model’s predicted distribution, which Acerbi-Szekely Z2 solves cleanly.

Why does Acerbi-Szekely Z2 outperform Z1 when VaR itself is miscalibrated?

Both Z1 and Z2 divide X_t by ES_t on days breaching VaR_t. However, Z1 divides by actual breaches N_T (which can mask parallel underestimation of both VaR and ES), whereas Z2 divides by expected breaches T*(1 - alpha), penalizing excess hit counts and severe losses together.

How do you obtain exact critical values for the Acerbi-Szekely Z2 test over 250 days?

Simulate M = 10,000 independent 250-day paths from the model’s own predictive return distribution F_t, compute Z2 on each path, and read the 5th and 1st empirical percentiles (typically around -0.61 and -0.92 under normality at 97.5% ES).

How does Fissler-Ziegel FZ0 scoring avoid unit-scale distortion across volatile and calm eras?

Because FZ0 is strictly 0-homogeneous—dividing forecast errors by negative ES and adding ln(-e)—high-volatility crisis weeks do not dominate the multi-year average score purely due to dollar scale, unlike quadratic scoring functions.

Publication record

Published
14 November 2026
Updated
14 November 2026
Author
Elena Rostova, FRM, CQF
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
3 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Elena Rostova, FRM, CQF. (2026, November 14). Acerbi-Szekely Expected Shortfall Backtesting: Non-Elicitability & Z2 Math. QuantCalc Tools. https://quantcalctools.com/guides/acerbi-szekely-expected-shortfall-backtesting-z2-statistic
Explore Parametric Var Expected Shortfall Hub (20) →

QuantCalc Tools Data, Benchmarks & Contributor Records

Acerbi-Szekely Expected Shortfall Z2 Backtesting Guide | QuantCalc Tools