关于「屈肌折痕:通贯掌的人群频率、卡方独立性与共同原因」,《手纹的图灵斑图与拓扑》给出的定量结论是什么?

结论:第四章用卡方与精确检验分析通贯掌与纹型的关联,并说明分层内独立的两种特征在合并后可以出现 1.393 的比值比。

数据口径:相学把它们当作写好的文书来读,本书则问三个可以计算的问题:这些纹怎样长出来?它们在人群中如何分布、彼此是否独立?从一枚成人指纹能否倒推胚胎期发生了什么?书中参数为示意,表格计数为合成数据,全部计算由书中程序生成;本书分析结构与概率,不对任何人的手作个人判断,也不是医疗、法律或经营建议。 第二章的二维有限差分模拟显示,一个系数的方向性就能把斑点变成平行嵴,细节点则由噪声决定。

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: 1.393

stratum share P(crease) P(whorl) odds ratio inside
low0.30 0.01 0.20 1.000
middle0.50 0.03 0.33 1.000
high0.20 0.08 0.50 1.000
pooled crease rate 0.0340, pooled whorl rate 0.3250
pooled odds ratio 1.393; Mantel-Haenszel 1.000
n1200: expected pooled chi-square 2.04 (df 1)
n5000: expected pooled chi-square 5.34 (df 1)
n20000: expected pooled chi-square 18.35 (df 1)
a pooled association that vanishes inside every stratum is
evidence of a shared cause, not of one trait reading another
查看完整验算程序
# Chapter 4. p42: A common cause makes two independent traits look
# linked. A latent hand geometry G (three strata, e.g. how high and wide
# the palm and finger pads are when creases and ridges form) raises both
# the chance of a single transverse crease and the chance of a whorl-
# dominant hand. Within each stratum the two traits are independent by
# construction. All probabilities are illustrative; the tables are
# exact expectations, not samples.
import math

STRATA = (            # share, P(crease | G), P(whorl-dominant | G)
    ("low", 0.30, 0.01, 0.20),
    ("middle", 0.50, 0.03, 0.33),
    ("high", 0.20, 0.08, 0.50),
)


def odds_ratio(a, b, c, d):
    return a * d / (b * c)


if __name__ == "__main__":
    cells = [[], [], [], []]
    print("  stratum   share  P(crease)  P(whorl)  odds ratio inside")
    mh_num, mh_den = [], []
    for name, w, pc, pw in STRATA:
        a = w * pc * pw
        b = w * pc * (1 - pw)
        c = w * (1 - pc) * pw
        d = w * (1 - pc) * (1 - pw)
        for k, val in enumerate((a, b, c, d)):
            cells[k].append(val)
        mh_num.append(a * d / w)
        mh_den.append(b * c / w)
        print("  {:<8s}  {:5.2f}  {:9.2f}  {:8.2f}  {:10.3f}".format(
            name, w, pc, pw, odds_ratio(a, b, c, d)))
    a, b, c, d = (math.fsum(x) for x in cells)
    print("pooled crease rate {:.4f}, pooled whorl rate {:.4f}".format(
        a + b, a + c))
    print("pooled odds ratio {:.3f}; Mantel-Haenszel {:.3f}".format(
        odds_ratio(a, b, c, d), math.fsum(mh_num) / math.fsum(mh_den)))
    pc, pw = a + b, a + c
    phi2 = (a * d - b * c) ** 2 / (pc * (1 - pc) * pw * (1 - pw))
    for n in (1200, 5000, 20000):
        print("n {:6d}: expected pooled chi-square {:7.2f} (df 1)".format(
            n, 1 + n * phi2))
    print("a pooled association that vanishes inside every stratum is")
    print("evidence of a shared cause, not of one trait reading another")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《手掌皮纹图灵反应扩散斑图、屈肌折痕拓扑分类与胚胎期应力反演》第 4 章「屈肌折痕:通贯掌的人群频率、卡方独立性与共同原因」(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-158 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026158,
  author    = {Fan, Yafu},
  title     = {Turing Reaction-Diffusion Patterns in Palm Dermatoglyphics, Flexion-Crease Topology and Embryonic Stress Inversion},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论