Delta-neutral MM

• ✓Runs in any market; steadier but smaller returns
• ⚠Spreads and funding can blow out in extreme conditions

Backtest curve is an example on a deterministic simulated price path — not real returns.

"""中性做市（Delta-neutral market making）—— 对称库存偏移 + 向中性衰减。

思路：与长仓做市（demos/market_making.py）同一库存偏移内核，但双向对称：价格高于滚动中间价做空、
低于中间价做多（等价于双边报价成交后的库存），并把目标仓位整体衰减 20% 使净 delta 持续回归 0——
赚取价差/波动回归而非方向。纯做市（pure market making）范式的独立重写（hummingbot 等开源做市
框架的同型思想）。
参数：ref_window（中间价窗口）、max_inventory（单边库存上限）、spread（价差敏感度）。

降级行为：连接器不支持做空时退化为长仓做市（目标裁剪到 [0, max_inventory]），docstring 即约定。
"""
from __future__ import annotations

from ..base import StrategyBase
from ..context import StrategyContext
from ..indicators import sma
from ..registry import register


@register("中性做市")
class DeltaNeutralMM(StrategyBase):
    description = "对称库存偏移做市：贵则做空、便宜则做多，目标向 0 衰减保持 delta 中性。"
    params = {"ref_window": 20, "max_inventory": 6.0, "spread": 0.02}

    def __init__(self, ref_window: int = 20, max_inventory: float = 6.0, spread: float = 0.02) -> None:
        self.ref_window = int(ref_window)
        self.max_inventory = float(max_inventory)
        self.spread = float(spread)

    async def on_tick(self, ctx: StrategyContext) -> None:
        sym = ctx.conn_symbol()
        hist = await ctx.history(sym, self.ref_window + 1)
        mid = sma(hist, self.ref_window)
        if mid is None or mid == 0:
            return
        dev = (hist[-1] - mid) / mid  # >0 偏贵, <0 偏便宜
        frac = max(-1.0, min(1.0, -dev / self.spread))
        if not getattr(ctx.conn, "allow_short", True):
            frac = max(0.0, frac)  # 长仓降级 = 现有做市 demo
        target = 0.8 * frac * self.max_inventory  # 衰减系数把净 delta 拉回中性
        await ctx.target(sym, target, band=self.max_inventory * 0.08)

The full SDK and all strategies are MIT-licensed open source — backtest, paper-trade, or fork them directly.