关于「季度合龙体检:流程、小工具与预注册证伪」,《族谱合龙与义庄永续》给出的定量结论是什么?
结论:第八章给出季度合龙体检、蒸馏小工具「孤岛家谱合龙匹配器」与预注册证伪方案,核心检验需核 385 对、每组盲核 435 对。书中以标注框区分已发表证据、可运行计算(20 段程序)、原典与作者假说。
数据口径:按字辈起名,同代人只剩一字可变,示意参数下一代 11 人就有一半机会撞名;两本 60 人对 40 人之支谱按全名硬接,接对者只占 0.208。 书中数据全为合成或示意,不含任何真实在世者资料;全部计算由书中程序生成;本书是非个人化分析,不构成法律、理财或经营建议,也不做任何血统高下或族群比较。
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: 385, 435
| H1 precision of the name key, reviewed same-name pairs | |
| half-width 0.10: n | 97 |
| half-width 0.05: n | 385 |
| half-width 0.03: n | 1,068 |
| H2 false merges among auto-links, per arm | |
| 0.20 vs 0.05: n | 76 |
| 0.10 vs 0.05: n | 435 |
| 0.20 vs 0.10: n | 199 |
| H3 founding lines extinct within ten generations | |
| at 0.8031, half-width 0.05: n | 243 |
| at 0.5, half-width 0.05 (worst case): n | 385 |
查看完整验算程序
# Chapter 8. p82: sample sizes for the pre-registered tests.
# H1 (Chapters 1 and 2): among pairs that share surname, generation
# character, given name and county, the share that are one person
# (precision of the name key) is at most 0.5. Estimated with a 95%
# interval of half-width E: n = z^2 p (1 - p) / E^2 at p = 0.5.
# H1 is refuted if the lower bound of that interval exceeds 0.9.
# H2: false-merge rate among auto-links, name key versus the
# two-threshold rule, planned at 0.20 versus 0.05 (alpha 0.05 two-sided,
# power 0.8), normal approximation for two proportions.
# H3 (Chapter 6): share of founding lines extinct within ten generations
# in consenting, de-identified books, planned near the Lotka value
# 0.8031 from p61, half-width 0.05.
# Planning values are the book's illustrative figures.
import math
from statistics import NormalDist
Z = NormalDist()
def n_one(p, half, conf=0.95):
z = Z.inv_cdf(0.5 + conf / 2.0)
return math.ceil(z * z * p * (1.0 - p) / (half * half))
def n_two(p1, p2, alpha=0.05, power=0.8):
za, zb = Z.inv_cdf(1.0 - alpha / 2.0), Z.inv_cdf(power)
pbar = (p1 + p2) / 2.0
top = (za * math.sqrt(2.0 * pbar * (1.0 - pbar)) +
zb * math.sqrt(p1 * (1.0 - p1) + p2 * (1.0 - p2))) ** 2
return math.ceil(top / (p1 - p2) ** 2)
if __name__ == "__main__":
print("H1 precision of the name key, reviewed same-name pairs")
for half in (0.10, 0.05, 0.03):
print(" half-width {:.2f}: n = {:,}".format(
half, n_one(0.5, half)))
print("H2 false merges among auto-links, per arm")
for p1, p2 in ((0.20, 0.05), (0.10, 0.05), (0.20, 0.10)):
print(" {:.2f} vs {:.2f}: n = {}".format(p1, p2, n_two(p1, p2)))
print("H3 founding lines extinct within ten generations")
print(" at 0.8031, half-width 0.05: n = {}".format(
n_one(0.8031, 0.05)))
print(" at 0.5, half-width 0.05 (worst case): n = {}".format(
n_one(0.5, 0.05)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《宗族世系有向无环图合龙算法、字辈同余编码与范氏义庄跨代治理母典》第 8 章「季度合龙体检:流程、小工具与预注册证伪」(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-334 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026334,
author = {Fan, Yafu},
title = {Genealogical DAG Merge Algorithms, Generation-Poem Congruence Coding and the Institutional Economics of the Fan Clan Charitable Estate},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}