Endogenous Hazard: The Model Starts Listening to the Market
Three articles ago I wrote that "endogeneity" was the deepest of the three critiques a derivatives-savvy reader made of my crash simulator. Two articles ago I built the margin spiral. Last article I built the dealer short-gamma spiral and ended with: "The last piece is to wire the gamma state into the hazard layer itself — that's V9-P2, and it's next." It's done. This is the closing of that loop.…
The hazard layer of a catastrophe-modeling framework previously operated independently, with historical crashes replaying the same way regardless of market conditions. However, the model has now been updated to incorporate market state into the hazard calculation, addressing a critique from Dean. The new system calculates event intensity using historical noise multiplied by a regime multiplier and a short-gamma factor.
The regime multiplier is determined by the VIX/SKEW state module, while the short-gamma factor comes from the dealer-gamma module. When the market is calm, the regime multiplier remains at 1.0, but the short-gamma term still causes historical crashes to replay about 20% harder due to dealers being net short gamma. This means that even in seemingly calm markets, there is an amplifier effect on crash replay intensity.
The model now acknowledges this feedback structure, demonstrating the endogeneity critique's central point: market contracts themselves create hazard. The mechanism is estimated, not directly observed, with some approximations made due to unavailable data. The next step involves incorporating discrete threshold effects into the model to better simulate forced selling and liquidation processes.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
