关于「数据处理不等式:体系再多也超不过出生时刻」,《十二体系的同源证据降权》给出的定量结论是什么?
结论:第六章用数据处理不等式说明,示意模型中十二套体系关于结果的信息恰为出生日本身的 0.5769 毫比特,一点不多;事后对上几乎必然,每句陈述只有 0.05 的机会对上时,十二套体系至少一句对上的概率是 0.9539。
数据口径:本书只分析符号体系的结构、信息量与概率,不做任何个人占断,不宣称任何体系有预测效力;书中参数为示意,全部计算由书中程序生成。 第六章用数据处理不等式说明,示意模型中十二套体系关于结果的信息恰为出生日本身的 0.5769 毫比特,一点不多;事后对上几乎必然,每句陈述只有 0.05 的机会对上时,十二套体系至少一句对上的概率是 0.9539。
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: 0.05, 0.9539
| statements fitting a past event after the fact | |||
| base rate F | systems | P(at least one fits) | expected fits |
| 0.05 | 1 | 0.2262 | 0.25 |
| 0.05 | 3 | 0.5367 | 0.75 |
| 0.05 | 12 | 0.9539 | 3.00 |
| 0.10 | 1 | 0.4095 | 0.50 |
| 0.10 | 3 | 0.7941 | 1.50 |
| 0.10 | 12 | 0.9982 | 6.00 |
| 0.30 | 1 | 0.8319 | 1.50 |
| 0.30 | 3 | 0.9953 | 4.50 |
| 0.30 | 12 | 1.0000 | 18.00 |
| familywise rate of 12 tests at 0.05, by effective count | |||
| independent | M_eff 12.00 | familywise 0.460 | Sidak level 0.00427 |
| row weights (p31) M_eff 9.23 familywise 0.377 Sidak level 0.00554 | |||
| participation (p31) M_eff 7.55 familywise 0.321 Sidak level 0.00677 | |||
| blocks (p52) M_eff 5.00 familywise 0.226 Sidak level 0.01021 | |||
查看完整验算程序
# Chapter 6. p62: matching after the fact, and testing many systems at once.
# Part 1: a reading offers S statements per system; any one statement
# fits a given past event with base rate F (a general statement fits
# many lives). Chance that at least one of M * S statements fits, and the
# expected number that fit. Part 2: familywise false-positive rate of M
# tests at level 0.05 when the tests are equivalent to M_eff independent
# ones (M_eff from p31 and p52), with the Sidak threshold. F, S
# illustrative.
import math
S = 5
print("statements fitting a past event after the fact")
print(" base rate F systems P(at least one fits) expected fits")
for f in (0.05, 0.1, 0.3):
for m in (1, 3, 12):
k = m * S
print("%12.2f %8d %21.4f %14.2f"
% (f, m, 1 - (1 - f) ** k, k * f))
print("familywise rate of 12 tests at 0.05, by effective count")
for name, me in (("independent", 12.0), ("row weights (p31)", 9.23),
("participation (p31)", 7.55), ("blocks (p52)", 5.0)):
fw = 1 - 0.95 ** me
sid = 1 - 0.95 ** (1 / me)
print(" %-20s M_eff %5.2f familywise %.3f Sidak level %.5f"
% (name, me, fw, sid))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《十二命理体系独立证据降权、同源生辰伪共振剥离与跨东西方符号系统正交合参潜变量母典》第 6 章「数据处理不等式:体系再多也超不过出生时刻」(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-283 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026283,
author = {Fan, Yafu},
title = {Down-Weighting Non-Independent Evidence across Twelve Symbolic Systems: Spurious Resonance from a Shared Birth Input and Orthogonal Latent-Variable Synthesis},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}