关于「集中流动性:放大因子、区间价值函数与出界概率」,《做市损耗与三明治上界》给出的定量结论是什么?

结论:第三章给出集中区间的放大因子,正负一成的区间在区间内放大到 21.49 倍,并计算出界概率。

数据口径:书中参数为示意,全部计算由书中程序生成。 本书是非个人化投资风险分析,不构成投资建议,不推荐任何代币、协议或交易;加密资产高风险,可能全部损失。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 21.49

range [1/r, r] around p1; sigma = 0.8 a year
half-width r A formula A numeric LVR/capital/yr
+-1 %1.01 201.50 201.50 16.1199
+-5 %1.05 41.49 41.49 3.3195
+-10 %1.10 21.49 21.49 1.7190
+-20 %1.20 11.48 11.48 0.9182
+-50 %1.50 5.45 5.45 0.4360
x22.00 3.41 3.41 0.2731
x44.00 2.00 2.00 0.1600
x100100.00 1.11 1.11 0.0889
off-centre: range [0.95, 1.20], price moves inside it
p0.96: A = 18.05
p1.00: A = 17.79
p1.05: A = 17.64
p1.10: A = 17.66
p1.19: A = 18.07
whitepaper reference: fee tiers 0.05 %, 0.30 %, 1 %; tick 1.0001
ticks from p to p r at r1.01: 100
ticks from p to p r at r1.10: 953
ticks from p to p r at r2.00: 6,932

核心公式

\[x(p) = L\left(\frac{1}{\sqrt{p}} - \frac{1}{\sqrt{p_b}}\right),\qquad y(p) = L\left(\sqrt{p} - \sqrt{p_a}\right),\]
查看完整验算程序
# Chapter 3. p31: concentrating liquidity multiplies LVR per unit capital.
# Model: a range position on [pa, pb] holds real reserves
# x = L (1/sqrt(p) - 1/sqrt(pb)), y = L (sqrt(p) - sqrt(pa)) while
# pa <= p <= pb, worth V = L (2 sqrt(p) - p / sqrt(pb) - sqrt(pa)).
# Inside the range |x'(p)| is the same as for a full-range pool with the
# same L, so the LVR rate (sigma^2 p^2 / 2) |x'(p)| is the same in money
# but larger per unit of capital. For a range [p / r, p r] centred on
# p the factor is A = 1 / (1 - r^(-1/2)). We check it with a numerical
# derivative. sigma = 0.8 a year; illustrative.
import math


def range_value(p, lq, pa, pb):
    p = min(max(p, pa), pb)
    return lq * (2.0 * math.sqrt(p) - p / math.sqrt(pb) - math.sqrt(pa))


def x_held(p, lq, pa, pb):
    p = min(max(p, pa), pb)
    return lq * (1.0 / math.sqrt(p) - 1.0 / math.sqrt(pb))


def lvr_per_capital(p, sigma, pa, pb, h=1e-6):
    lq = 1.0
    dx = (x_held(p + h, lq, pa, pb) - x_held(p - h, lq, pa, pb)) / (2 * h)
    rate = sigma ** 2 * p * p / 2.0 * abs(dx)
    return rate / range_value(p, lq, pa, pb)


if __name__ == "__main__":
    sigma = 0.8
    base = sigma ** 2 / 8.0
    print("range [1/r, r] around p = 1; sigma = 0.8 a year")
    print("  half-width   r      A formula  A numeric  LVR/capital/yr")
    for label, r in (("+-1 %", 1.01), ("+-5 %", 1.05), ("+-10 %", 1.10),
                     ("+-20 %", 1.20), ("+-50 %", 1.50), ("x2", 2.0),
                     ("x4", 4.0), ("x100", 100.0)):
        a = 1.0 / (1.0 - r ** -0.5)
        num = lvr_per_capital(1.0, sigma, 1.0 / r, r) / base
        print("  {:<9} {:>6.2f}  {:>9.2f}  {:>9.2f}  {:>13.4f}".format(
            label, r, a, num, a * base))
    print("off-centre: range [0.95, 1.20], price moves inside it")
    for p in (0.96, 1.0, 1.05, 1.1, 1.19):
        print("  p = {:.2f}: A = {:>6.2f}".format(
            p, lvr_per_capital(p, sigma, 0.95, 1.20) / base))
    print("whitepaper reference: fee tiers 0.05 %, 0.30 %, 1 %; tick 1.0001")
    for r in (1.01, 1.10, 2.0):
        print("  ticks from p to p r at r = {:.2f}: {:,.0f}".format(
            r, math.log(r) / math.log(1.0001)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《集中流动性做市商重平衡损耗(LVR)、链上三明治抢跑边界与去中心化金库风控母典》第 3 章「集中流动性:放大因子、区间价值函数与出界概率」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-344 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026344,
  author    = {Fan, Yafu},
  title     = {Concentrated-Liquidity Loss-Versus-Rebalancing, On-Chain Sandwich Bounds and Decentralised Vault Risk Control},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论