关于「方差风险溢价:看跌多头为什么平均是负的」,《低波动窗口的看跌期权》给出的定量结论是什么?
结论:第一章用 Black–Scholes 希腊值与方差风险溢价说明期望与胜率是两回事:假设指数每年下跌一成,平值看跌期权持有到期的胜率也只有 0.4358。
数据口径:书中的「做空」只指损失有上限的看跌敞口,即买入看跌期权与看跌借方价差;全部价格以合成指数的指数点表示,不使用任何历史行情。 书中参数为示意;全部计算由书中程序生成。
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: 0.4358
| 30-day put, iv 0.14, realised 0.13, held to expiry | ||||
| K mu exact MC Wilson 95 % E P&L/prem | ||||
| 4,000 +0.06 | 0.3035 | 0.3058 | [0.2994, 0.3122] | -0.148 |
| 4,000 -0.10 | 0.4358 | 0.4365 | [0.4296, 0.4433] | +0.303 |
| 4,000 -0.30 | 0.6100 | 0.6077 | [0.6009, 0.6144] | +1.023 |
| 3,800 +0.06 | 0.0622 | 0.0640 | [0.0607, 0.0675] | -0.346 |
| 3,800 -0.10 | 0.1183 | 0.1179 | [0.1136, 0.1225] | +0.369 |
| 3,800 -0.30 | 0.2289 | 0.2299 | [0.2241, 0.2358] | +2.073 |
查看完整验算程序
# Chapter 1. p13: the thirty-day hit rate of a long put is far below one.
# A 30-day put bought at implied vol 0.14 is held to expiry; a "hit" is
# a payoff above the premium carried at r. Index lognormal under the
# real-world measure with realised vol 0.13 and drift mu. The exact
# hit probability P(S_T < K - premium * exp(rT)) is compared with
# 20,000 Monte Carlo draws and their Wilson 95 % interval (Wilson 1927).
# Synthetic, illustrative parameters; no market data.
import math
import random
S, R, T, IV, SR = 4000.0, 0.03, 30.0 / 365.0, 0.14, 0.13
N = 20000
def ncdf(x):
return 0.5 * (1.0 + math.erf(x / math.sqrt(2.0)))
def put_price(s, k, t, r, sig):
sq = sig * math.sqrt(t)
d1 = (math.log(s / k) + (r + 0.5 * sig * sig) * t) / sq
return k * math.exp(-r * t) * ncdf(-(d1 - sq)) - s * ncdf(-d1)
def wilson(h, n, z=1.959964):
p = h / n
den = 1.0 + z * z / n
mid = (p + z * z / (2.0 * n)) / den
half = z * math.sqrt(p * (1.0 - p) / n + z * z / (4.0 * n * n)) / den
return mid - half, mid + half
rng = random.Random(42513)
zs = [rng.gauss(0.0, 1.0) for _ in range(N)]
print("30-day put, iv 0.14, realised 0.13, held to expiry")
print(" K mu exact MC Wilson 95 % E P&L/prem")
for k in (4000.0, 3800.0):
cost = put_price(S, k, T, R, IV) * math.exp(R * T)
for mu in (0.06, -0.10, -0.30):
m = (mu - 0.5 * SR * SR) * T
sd = SR * math.sqrt(T)
exact = ncdf((math.log((k - cost) / S) - m) / sd)
pays = [max(k - S * math.exp(m + sd * z), 0.0) for z in zs]
hits = len([x for x in pays if x > cost])
lo, hi = wilson(hits, N)
ev = math.fsum(pays) / N / cost - 1.0
print("{:>5,} {:+.2f} {:.4f} {:.4f} [{:.4f}, {:.4f}] {:+.3f}"
.format(int(k), mu, exact, hits / N, lo, hi, ev))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《宏观下行期低隐含波动率做空窗口、双希腊值凸性共振与三十日胜率边界母典》第 1 章「方差风险溢价:看跌多头为什么平均是负的」(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-425 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026425,
author = {Fan, Yafu},
title = {Shorting Low-Implied-Volatility Windows in Macro Downturns: Delta-Vega Dual Convexity, Bear-Rally Compression and Thirty-Day Hit-Rate Bounds},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}