关于「东西方原典考辨:贵上极则反贱、投机者的零期望与偿付能力引语」,《低波动窗口的看跌期权》给出的定量结论是什么?
结论:第七章考辨《史记·货殖列传》「贵上极则反贱」、巴舍利耶「投机者的数学期望为零」、凯利关于信息率的原话,以及一句被误归于凯恩斯的格言;从低位出发,先回到中位数与先跌得更低的概率是 0.5095 对 0.4586。
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: 0.4586, 0.5095
| start iv 0.1259 (5th pct), median 0.19, 2nd pct 0.1137 | ||||
| day | median iv | P(iv>0.19) | P(iv<start) | P(iv>0.25) |
| 10 | 0.1345 | 0.0053 | 0.3143 | 0.0000 |
| 30 | 0.1488 | 0.1119 | 0.2035 | 0.0049 |
| 60 | 0.1643 | 0.2671 | 0.1278 | 0.0363 |
| 90 | 0.1743 | 0.3618 | 0.0920 | 0.0699 |
| race over 120 days: median first 0.5095, 2nd pct first 0.4586, | ||||
| neither 0.0319; days to median, quartiles 26 / 40 / 61 |
查看完整验算程序
# Chapter 7. p71: "what is extremely cheap turns dear" as a mean-
# reverting log implied vol. x = ln(iv) is Ornstein-Uhlenbeck with
# median iv 0.19, stationary sd 0.25 and a half-life of 40 days. It
# starts at its 5th stationary percentile. Exact Gaussian transitions
# give the day-t distribution; a 10,000-path exact daily simulation
# gives the race between touching the median and first sinking to the
# 2nd percentile. Illustrative parameters, synthetic paths only.
import math
import random
from statistics import NormalDist
ND = NormalDist()
M, SD, HL = math.log(0.19), 0.25, 40.0
KAP = math.log(2.0) / HL
X0 = M + ND.inv_cdf(0.05) * SD
LOW = M + ND.inv_cdf(0.02) * SD
print("start iv {:.4f} (5th pct), median 0.19, 2nd pct {:.4f}".format(
math.exp(X0), math.exp(LOW)))
print("day median iv P(iv>0.19) P(iv<start) P(iv>0.25)")
for t in (10, 30, 60, 90):
mu = M + (X0 - M) * math.exp(-KAP * t)
sd = SD * math.sqrt(1.0 - math.exp(-2.0 * KAP * t))
d = NormalDist(mu, sd)
print(" {:>2} {:.4f} {:.4f} {:.4f} {:.4f}".format(
t, math.exp(mu), 1.0 - d.cdf(M), d.cdf(X0),
1.0 - d.cdf(math.log(0.25))))
rng = random.Random(42571)
a = math.exp(-KAP)
s1 = SD * math.sqrt(1.0 - a * a)
up = down = 0
first_up = []
for _ in range(10000):
x = X0
for day in range(1, 121):
x = M + (x - M) * a + s1 * rng.gauss(0.0, 1.0)
if x >= M:
up += 1
first_up.append(day)
break
if x <= LOW:
down += 1
break
first_up.sort()
print("race over 120 days: median first {:.4f}, 2nd pct first {:.4f},"
.format(up / 10000, down / 10000))
print("neither {:.4f}; days to median, quartiles {} / {} / {}".format(
1.0 - (up + down) / 10000, first_up[len(first_up) // 4],
first_up[len(first_up) // 2], first_up[3 * len(first_up) // 4]))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 7 章「东西方原典考辨:贵上极则反贱、投机者的零期望与偿付能力引语」(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}
}