Volatility Perpetuals (VXX)

This perpetual contract tracks the short-term VIX futures roll, providing exposure to the implied volatility of the U.S. stock market.

Characteristic	Description
What it Tracks	A blend of the front two VIX futures contracts (VX1/VX2) with a daily roll weight λₜ.
Index Definition	Xt = (1-λt)m₁(t) + λt m₂(t) where m₁ and m₂ are the mid-quotes of VX1 and VX2.
Oracle Type	A Session-Selected Oracle that uses different methodologies based on underlying market hours.
Market Type	Cash-Settled Index Perpetual.
PnL Formula	PNL (Long) = (Exit_Price - Entry_Price) × Position_Size + Total_Funding_Received PNL (Short) = (Entry_Price - Exit_Price) × Position_Size + Total_Funding_Received Where: Entry_Price / Exit_Price: The price at which a position is opened and closed, respectively. Position_Size: The number of VXXN contracts in a position. Total_Funding_Received (or Total_Funding_Paid): The sum of all hourly funding payments accrued over the duration of the position. This value can be positive (if a user were paid by the other side) or negative (if the user paid the other side).
Use Cases	Long (Profit from rising fear): • A direct, capital-efficient hedge against broad market downturns. • A speculative tool to position for volatility spikes around major economic events (e.g., CPI, FOMC). Short (Earn the volatility premium): • Earn the "yield" generated from volatility's natural decay during calm or bullish markets. • A way to express a view that market fear is overpriced.
Users	• Portfolio Managers & DAOs (Hedgers) • Macro Traders & Event-Driven Funds (Speculators) • Sophisticated Retail Traders
Vol-stat (σ)	High: the VXXN is a high-volatility product by design and can experience double-digit percentage moves in a single day. Risk parameters are set accordingly.
Margin Numbers	Derived from the Max Leverage tiers. For the initial 3x leverage tier, the Initial Margin (IM) is ~33.3% of the notional position value.

OraclePx

This is run for the VXX SEDA feed. The VXXN-PERP oracle is session-aware, ensuring a robust price feed 24/7.

• During CFE Market Hours: The oracle directly tracks the VIX futures replica: Mark Price ← X_t.

• During Off-Hours: When the CFE is closed, the oracle uses a synthetic model to infer the price of VX1/VX2 based on actively trading proxies like VSTOXX futures (FVS) or E-mini S&P futures (ES).

• During Weekends: The oracle uses a low-drift, state-space model to prevent erratic price movements while maintaining a live feed. This is a low-volatility state-space model:

with ρΦ<1 and |Σ| scaled for drift ≤ 25 bps/hr.

This requires explicit governance enable flag. Apply stricter clamps than weekdays.

The final oracle price (oraclePx) is a composite rate produced by SEDA, which blends data from Pyth with Nunchi's on-chain mark price from Hyperliquid. SEDA applies session-based weights, favoring the live Hyperliquid mark during active trading and leaning more heavily on the Pyth reference price during off-market hours to ensure stability.

MarkPx

This is a feed based on our market. The markPx is calculated as a median of three variants (MarkPx0, MarkPx1, MarkPx2) to protect against manipulation,

ExternalPerpPx Pricing

This is the VXX SEDA feed. The externalPerpPx is set to the SEDA feed to be used as a band-limiter.

Worked Example

This example demonstrates how the mark price is calculated during a session switch, when the primary market (CFE) closes and the oracle begins inferring prices from an off-hours proxy.

Suppose the CFE closes at time t* with the front and second-month VIX futures at (m₁, m₂) = (17.00, 20.50) and the daily roll weight λ = 0.30.

The EU session begins, and over the next 2 hours, the Eurex VSTOXX futures (FVS) gain 2%. The calibrated log-return betas for FVS are β_FVS⁽¹⁾ = 0.60 and β_FVS⁽²⁾ = 0.55.

First, we infer the synthetic prices for the VIX futures using the off-hours inference formula (8):

• Inferred Front-Month Future (m̂₁):

  • m̂₁ ≈ 17.00 * exp(0.60 * 0.02) ≈ 17.00 * exp(0.012) ≈ 17.21

• Inferred Second-Month Future (m̂₂):

  • m̂₂ ≈ 20.50 * exp(0.55 * 0.02) ≈ 20.50 * exp(0.011) ≈ 20.73

Next, we calculate the new mark price M using the VIX roll replica formula (9) with the inferred future prices:

M ≈ (1 - 0.30) * 17.21 + 0.30 * 20.73

M ≈ 0.7 * 17.21 + 0.3 * 20.73 ≈ 12.05 + 6.22 ≈ 18.27

This new price of ~$18.27 would be the published mark, subject to the standard 1% per-update clamp and quantization rules.