关于「实证闭环:排盘服务的高纬防错补丁、季节解释卡与预注册检验」,《宫位几何与出生季节》给出的定量结论是什么?

结论:第八章给出排盘服务的高纬防错补丁、出生季节解释卡、蒸馏小工具「高纬宫位校验器」与两项预注册检验:三选一配对检验在真实正确率 0.45 时需要 111 张星盘,截止日平移检验每个协会需要 512 名入选者。书中以标注框区分已发表的证据、可运行的 Python 计算(19 段程序)、原典与作者的假说。

数据口径:书中参数为示意;全部计算由书中程序生成。 书中以标注框区分已发表的证据、可运行的 Python 计算(19 段程序)、原典与作者的假说。

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: 0.45, 111, 512

Test A: three-choice matching, H0 hit rate 0.3333
  true rate   charts n   hits needed
       0.40        325           123
       0.45        111            46
       0.50         58            26
Test B: share born Jan-Mar among selected players
  cut-off 1 Jan 0.3141, cut-off 1 Aug 0.2359
  players needed per federation: 512
  sun-sign account predicts equal shares in both
查看完整验算程序
# Chapter 8. p82: sample sizes for the two pre-registered tests.
# Test A, a Carlson-style matching test: each chart comes with three
# profiles, one genuine; H0 hit rate 1/3. Exact one-sided binomial
# test, alpha 0.05, power 0.80, for alternatives 0.40, 0.45, 0.50 (the
# last is the minimum hit rate astrologers forecast in Carlson 1985).
# Test B, the cut-off-shift test: selected players from a federation
# with a 1 January cut-off and one with a 1 August cut-off; outcome =
# share born January to March. Cut-off account: shares from the model
# of p12 (beta 0.4, top 0.10, illustrative); sun-sign account: equal
# shares. Two-sided two-proportion z-test, alpha 0.05, power 0.80.
# 模型:两项预注册检验的样本量;效应量为示意值。
import math

DAYS = (31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31)


def binom_tail(n, k, p):
    if k <= 0:
        return 1.0
    terms = []
    for j in range(k, n + 1):
        lt = (math.lgamma(n + 1) - math.lgamma(j + 1) - math.lgamma(n - j + 1)
              + j * math.log(p) + (n - j) * math.log(1.0 - p))
        terms.append(math.exp(lt))
    return math.fsum(terms)


def exact_n(p0, p1, alpha=0.05, target=0.80):
    for n in range(10, 3000):
        k = next(j for j in range(n + 2)
                 if binom_tail(n, j, p0) <= alpha)
        if binom_tail(n, k, p1) >= target:
            return n, k
    return None, None


def q(x):
    return 0.5 * math.erfc(x / math.sqrt(2.0))


def jan_mar_share(cut, beta=0.4, s=0.10):
    ages = [1.0 - (((d - cut) % 365) + 0.5) / 365.0 for d in range(365)]
    lo, hi = -10.0, 10.0
    for _ in range(100):
        c = 0.5 * (lo + hi)
        if math.fsum(q(c - beta * a) for a in ages) / 365.0 > s:
            lo = c
        else:
            hi = c
    w = [q(c - beta * a) for a in ages]
    return math.fsum(w[0:90]) / math.fsum(w)


def two_prop_n(p1, p2):
    za, zb = 1.959964, 0.841621
    pb = 0.5 * (p1 + p2)
    a = za * math.sqrt(2.0 * pb * (1.0 - pb))
    b = zb * math.sqrt(p1 * (1.0 - p1) + p2 * (1.0 - p2))
    return math.ceil(((a + b) / (p1 - p2)) ** 2)


if __name__ == "__main__":
    print("Test A: three-choice matching, H0 hit rate 0.3333")
    print("  true rate   charts n   hits needed")
    for p1 in (0.40, 0.45, 0.50):
        n, k = exact_n(1.0 / 3.0, p1)
        print("  {:9.2f}   {:8d}   {:11d}".format(p1, n, k))
    pa = jan_mar_share(0)
    pb = jan_mar_share(int(math.fsum(DAYS[:7])))
    print("Test B: share born Jan-Mar among selected players")
    print("  cut-off 1 Jan {:.4f}, cut-off 1 Aug {:.4f}".format(pa, pb))
    print("  players needed per federation: {}".format(two_prop_n(pa, pb)))
    print("  sun-sign account predicts equal shares in both")

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《西洋本命盘宫位制球面几何、极圈奇点退化与出生季节效应的非占星解释母典》第 8 章「实证闭环:排盘服务的高纬防错补丁、季节解释卡与预注册检验」(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-227 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026227,
  author    = {Fan, Yafu},
  title     = {Spherical Geometry of Western House Systems, Polar-Circle Singularities and Non-Astrological Explanations of Birth-Season Effects},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论