Quantitative#096

Gaussian-Process Surface for Sparse Strike Markets

Threshold markets on the same theme (e.g. BTC above various prices and dates) are often priced thinly and inconsistently. A Gaussian process is a method that draws one smooth fair-value surface across the whole (price, time) grid AND reports how confident it is at each point. You trade any market sitting off that surface by more than the model's uncertainty allows. The edge is borrowing pricing information from related markets to value each thin one more accurately than its own sparse book can.

What you need to run it

  • Cross-market quotes across strikes/expiries in a theme
  • GP regression with kernel tuned to event geometry
  • Uncertainty-aware trade triggers (posterior mean +/- sigma)
  • Execution across multiple illiquid nodes with size limits

Where this applies

Markets on Polymarket where gaussian-process surface for sparse strike markets is the natural play:

  • Will Bitcoin close above $175k on Dec 31, 2026? (node on the surface)
  • Will Ethereum exceed $8,000 by Mar 31, 2027?
  • Will Solana close above $400 on Sept 30, 2026?

Capabilities this demands

Model / quantCustom code / APIData ingestionMulti-venue

At a glance

CategoryQuantitative
Requirements4
CapabilitiesModel / quant, Custom code / API, Data ingestion, Multi-venue
VenuePolymarket (CLOB, Polygon)

Build it

Related quantitative strategies

This is documentation, not advice. Poly Research & Robotics publishes how these strategies work because the method should be checkable — not as a recommendation to trade them. See the full strategy database (147 strategies) or the data resources directory.
Join Discord