关于「条文的特异度与刻分的分辨率:巴纳姆句、同分钟出生者与双胞胎」,《铁板神数的比特账》给出的定量结论是什么?

结论:第五章说明一句条文的期望信息量至多 0.5307 比特,按示意规模每分钟有 19.01 人出生。

数据口径:书中参数为示意;全部计算由书中程序生成;本书分析结构与概率,不对任何个人作任何占断,也不构成法律或理财建议。 第五章说明一句条文的期望信息量至多 0.5307 比特,按示意规模每分钟有 19.01 人出生。

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: 0.5307

verse share p bits p*bits
worked hard when young0.8500 0.234 0.199
one parent stricter than the other0.8000 0.322 0.258
has at least one sibling0.7000 0.515 0.360
both parents living0.6000 0.737 0.442
exactly two brothers0.2000 2.322 0.464
father born in a Rat year0.0833 3.585 0.299
father Rat, mother Ox0.0069 7.170 0.050
best expected surprisal: p0.3679, 0.5307 bits per verse
reading of 12 verses: expected confirmed 6.8 (0.567)
chance at least one specific verse is confirmed0.3439
expected bits carried by confirmed verses3.389
Forer 1949: mean rating 4.26 of 5 = share 0.852
查看完整验算程序
# Chapter 5. p51: The trade between being confirmed and being
# informative. A verse true of a share p of clients is confirmed with
# probability p and, when confirmed, carries -log2 p bits. Its expected
# surprisal p * (-log2 p) peaks at p = 1/e. Shares are illustrative.
# Forer (1949) reports a mean accuracy rating of 4.26 on a 0 to 5 scale
# for one generic sketch given to every student; we print its share.
import math

verses = [
    ("worked hard when young", 0.85),
    ("one parent stricter than the other", 0.80),
    ("has at least one sibling", 0.70),
    ("both parents living", 0.60),
    ("exactly two brothers", 0.20),
    ("father born in a Rat year", 1 / 12),
    ("father Rat, mother Ox", 1 / 144),
]
print("  verse                                share p   bits   p*bits")
for name, p in verses:
    b = -math.log2(p)
    print("  {:<36}{:7.4f}  {:6.3f}  {:6.3f}".format(name, p, b, p * b))
pk = 1 / math.e
print("  best expected surprisal: p = {:.4f}, {:.4f} bits per verse"
      .format(pk, -pk * math.log2(pk)))

# a reading of 8 generic verses (p = 0.8) and 4 specific ones (p = 0.1)
ps = [0.8] * 8 + [0.1] * 4
conf = math.fsum(ps)
print("  reading of 12 verses: expected confirmed {:.1f} ({:.3f})".format(
    conf, conf / 12))
print("  chance at least one specific verse is confirmed: {:.4f}".format(
    1 - 0.9 ** 4))
info = math.fsum(-p * math.log2(p) for p in ps)
print("  expected bits carried by confirmed verses: {:.3f}".format(info))
FORER_MEAN = 4.26
print("  Forer 1949: mean rating {} of 5 = share {:.3f}".format(
    FORER_MEAN, FORER_MEAN / 5))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《铁板神数考刻分决策树、条文库仿射编码与信息熵套取下界》第 5 章「条文的特异度与刻分的分辨率:巴纳姆句、同分钟出生者与双胞胎」(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-157 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026157,
  author    = {Fan, Yafu},
  title     = {Tieban Shenshu Ke-Fen Decision Trees, Affine Coding of the Verse Bank and Entropy Lower Bounds on Elicitation},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论