Adaptive Grid Trading with Recursive Trend and Inventory Controls
Summary
The article redesigns a leveraged grid strategy after identifying low activity in quiet markets, accumulating exposure during sustained declines, profit giveback, and stubborn high-priced inventory. It frames grids as short volatility strategies whose oscillation gains can be overwhelmed by growing directional inventory in a trend. The proposed system varies grid width with an online estimate of equilibrium dispersion, using a local level and slope filter and a recursive autoregressive model to estimate mean-reversion speed and half-life. A timeout based on that estimated half-life can trigger an exit when reversion takes too long.
Additional safeguards estimate systematic movement from related assets, block countertrend entries when those assets confirm the move, cap net exposure, lower the effective ceiling as long inventory grows, and halt new entries after drawdown limits. Synthetic examples illustrate how reference assets can distinguish synchronized declines from target-specific movement. The article does not establish profitability: it calls for backtesting, and its mathematical claims and implementation are not independently validated.
Key ideas
- A grid earns from oscillation but can accumulate dangerous directional inventory during persistent trends.
- An online level and slope filter estimates trend strength, while recursive regression estimates mean-reversion scale and half-life.
- Grid width adapts to estimated equilibrium dispersion rather than relying only on a fixed interval.
- Related assets help distinguish broad market drift from target-specific movement.
- Exposure caps, inventory skew, trend gates, and drawdown controls restrict new entries.
- Synthetic illustrations do not establish profitability, which remains subject to backtesting.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.