关于「分位数与期望短缺:怎样量度经营资本的一年尾部」,《尾部保护与凯利边界》给出的定量结论是什么?
结论:把指数换成经营者的资本 \(M = 1 + A + L(S - 1)\) 之后,程序 p31 显示:在默顿模型下,97.5 % 的期望短缺是 0.3373,比同均值同方差的对数正态的 0.2588 深得多;在 99.5 % 的水平上,期望短缺是 0.4493。在正态世界里,99 % 的在险价值 2.3263 与 97.5 % 的期望短缺 2.3378 个标准差几乎相等。程序 p32 显示两家各自 95 % 在险价值为 -2.00 的分店,合起来的在险价值是 98.00,违反次可加性;期望短缺合起来是 101.26,小于分开之和 159.20。程序 p33 显示 20 年记录估出的期望短缺平均只有 0.2266,有 0.601 的机会低估超过四分之一。下一章回答一个随之而来的问题:既然尾部难以估准、保护又按风险溢价收费,有限的预算应该先保住尾巴的哪一段?
数据口径:用默顿跳扩散模型描述营收指数,并让保费报价对跳跃风险收取示意的溢价,行权 0.80 的固定保费保护,保费是平均赔付的 1.669 倍,买方平均收益为 -0.401;越只保极端情形,保费相对平均赔付越贵。 书中参数为示意;全部计算由书中程序生成;本书是非个人化法律、税务或理财建议,不推荐任何具体金融产品、保险产品、机构或交易策略,不承诺收益,不鼓励杠杆。
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: 0.2588, 0.3373, 0.4493, 2.3263, 2.3378
| capital loss as a share of start capital (A = 0.12, L = 1.0) | |||||
| level | VaR Merton | ES Merton | | | VaR lognorm | ES lognorm |
| 0.950 | 0.1744 | 0.2736 | | | 0.1601 | 0.2204 |
| 0.975 | 0.2531 | 0.3373 | | | 0.2076 | 0.2588 |
| 0.990 | 0.3390 | 0.4059 | | | 0.2589 | 0.3016 |
| 0.995 | 0.3905 | 0.4493 | | | 0.2916 | 0.3294 |
| normal check: VaR 0.99 = 2.3263 sd, ES 0.975 = 2.3378 sd |
查看完整验算程序
# Chapter 3. p31: value-at-risk and expected shortfall of a year's
# capital loss for an operator, exact under Merton and under a lognormal
# with the same mean and variance of log S.
# Year-end capital per unit of start capital: M = 1 + A + L * (S - 1),
# A = planned return on capital, L = exposure of capital to the revenue
# index S. Loss = 1 - M. VaR_a is the a-quantile of the loss and
# ES_a the mean loss beyond it. Parameters illustrative.
import math
MU, SIG = 0.05, 0.18
LAM, JM, JS = 0.10, -0.35, 0.15
A, L = 0.12, 1.0 # 计划资本回报与营收敞口(示意)
def Phi(x):
return 0.5 * math.erfc(-x / math.sqrt(2.0))
def phi(x):
return math.exp(-0.5 * x * x) / math.sqrt(2.0 * math.pi)
def merton():
kap = math.exp(JM + 0.5 * JS * JS) - 1.0
base = MU - 0.5 * SIG * SIG - LAM * kap
out = []
for n in range(13):
w = math.exp(-LAM) * LAM ** n / math.factorial(n)
out.append((w, base + n * JM, math.sqrt(SIG * SIG + n * JS * JS)))
return out
def matched(cs):
mean = math.fsum(w * m for w, m, s in cs)
var = math.fsum(w * (s * s + m * m) for w, m, s in cs) - mean * mean
return [(1.0, mean, math.sqrt(var))]
def cdf(cs, x):
return math.fsum(w * Phi((math.log(x) - m) / s) for w, m, s in cs)
def partial(cs, x):
# E[S; S < x]
return math.fsum(w * math.exp(m + 0.5 * s * s)
* Phi((math.log(x) - m) / s - s) for w, m, s in cs)
def var_es(cs, a):
lo, hi = 1e-6, 5.0
for _ in range(100):
mid = 0.5 * (lo + hi)
if cdf(cs, mid) < 1.0 - a:
lo = mid
else:
hi = mid
q = 0.5 * (lo + hi)
es_s = partial(cs, q) / (1.0 - a)
return 1.0 - (1.0 + A + L * (q - 1.0)), 1.0 - (1.0 + A + L * (es_s - 1.0))
cm = merton()
cl = matched(cm)
print("capital loss as a share of start capital (A = 0.12, L = 1.0)")
print("level VaR Merton ES Merton | VaR lognorm ES lognorm")
for a in (0.95, 0.975, 0.99, 0.995):
v1, e1 = var_es(cm, a)
v2, e2 = var_es(cl, a)
print("{:.3f} {:.4f} {:.4f} | {:.4f} {:.4f}".format(
a, v1, e1, v2, e2))
z99 = 2.3263478740
z975 = 1.9599639845
print("normal check: VaR 0.99 = {:.4f} sd, ES 0.975 = {:.4f} sd".format(
z99, phi(z975) / 0.025))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《尾部黑天鹅期权护城河与分位数对冲、跳扩散模型与凯利边界下的固定保费现金流保护母典》第 3 章「分位数与期望短缺:怎样量度经营资本的一年尾部」(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-290 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026290,
author = {Fan, Yafu},
title = {Tail-Risk Option Moats and Quantile Hedging: Jump-Diffusion Models and Fixed-Premium Protection of Core Cash Flows under Kelly Bounds},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}