关于「东西方原典考辨:曾参杀人与大森余震律」,《评论裂变分支比》给出的定量结论是什么?

结论:第七章考辨《战国策》曾参杀人(分支比 0.8 时三次告知同源的概率 0.6400)与大森房吉的余震律(伽马混合下 p 为 1.4)。

数据口径:书中的帖子、账号、流量、成本与评论价值均为示意参数;全部计算由书中程序生成;本书是非个人化经营建议,只讨论真实用户的评论与平台官方工具买来的曝光,不涉及任何制造评论或虚构稀缺的手段。 第四章用小崎递推、补偿子恒等式与黄金分割求最大似然估计,示意帖的分支比 95 % 区间从 0.42 到 0.92,两小时估计的标准差约 0.186。

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: 0.6400, 0.8

reports naive post. n0.5 n=0.8 n=0.9 P(one source, n=0.8)
10.0099 0.0099 0.0099 0.0099 1.0000
20.0910 0.0307 0.0156 0.0124 0.8000
30.5003 0.0910 0.0245 0.0156 0.6400
50.9901 0.5003 0.0594 0.0245 0.4096
101.0000 0.9969 0.3871 0.0737 0.1342
201.0000 1.0000 0.9844 0.4429 0.0144
n0.0: reports needed to pass even odds: 3
n0.5: reports needed to pass even odds: 5
n0.8: reports needed to pass even odds: 11
n0.9: reports needed to pass even odds: 21
independent voices in 260 comments at n0.6: about 104
查看完整验算程序
# Chapter 7. p72: Zeng Shen's mother: three reports, or one report told
# three times? Bayes with echoes.
# Prior odds that the son killed a man: 1 to 999 (illustrative). An
# independent eyewitness report carries a likelihood ratio L = 10; a
# retelling of an earlier report carries none (L = 1). In a cascade
# with branching ratio n, each later report is a retelling with
# probability n, so the expected log evidence after k reports is
# log L (1 + (k - 1)(1 - n)). Also: the chance that all k reports come
# from a single source, n^(k - 1).
import math

PRIOR, L = 1.0 / 999.0, 10.0


def posterior(k, n, naive):
    indep = k if naive else 1 + (k - 1) * (1 - n)
    odds = PRIOR * L ** indep
    return odds / (1 + odds)


print("reports  naive post.  n=0.5   n=0.8   n=0.9   P(one source, n=0.8)")
for k in (1, 2, 3, 5, 10, 20):
    print("%4d     %.4f       %.4f  %.4f  %.4f  %.4f"
          % (k, posterior(k, 0, True), posterior(k, 0.5, False),
             posterior(k, 0.8, False), posterior(k, 0.9, False),
             0.8 ** (k - 1)))
for n in (0.0, 0.5, 0.8, 0.9):
    k = 1
    while posterior(k, n, False) < 0.5:
        k += 1
    print("n = %.1f: reports needed to pass even odds: %d" % (n, k))
print("independent voices in 260 comments at n = 0.6: about %.0f"
      % (260 * (1 - 0.6)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《霍克斯自激点过程、评论裂变分支比临界值与直播跟风抢购余震半衰期母典》第 7 章「东西方原典考辨:曾参杀人与大森余震律」(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-234 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026234,
  author    = {Fan, Yafu},
  title     = {Hawkes Self-Exciting Point Processes: Branching-Ratio Criticality in Comment Cascades and the Aftershock Half-Life of Livestream Herd Buying},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论