关于「实践:样盘 A 的解读卡,以及一份对照巴纳姆的盲配检验」,《子平调候与十神博弈》给出的定量结论是什么?

结论:第八章用一张不含任何出生资料的样盘示范查法:调候格不受时辰影响,主导十神在十二个时辰里都不变,组合一层最脆弱;并给出一份与巴纳姆对照臂并行、预先登记的盲配检验方案:三选一、真实命中率 0.40 时需要 325 人,两个主要终点时需要 415 人。书中以标注框区分已发表的证据、可运行的 Python 计算、原典与作者的假说。

数据口径:全部计算由书中程序生成;本书不是投资、职业、心理、法律或婚姻建议。 书中以标注框区分已发表的证据、可运行的 Python 计算、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: 0.40, 325, 415

arm 1 k chance true hit N (one end) N (two ends)
30.3333 0.40 325 415
30.3333 0.45 111 139
30.3333 0.50 58 68
50.2000 0.30 116 144
50.2000 0.35 56 72
N300, k = 3: reject if hits >= 115; false-positive rate 0.0390
arm 2 own card minus Barnum profile, dz0.15: N = 349
arm 2 own card minus Barnum profile, dz0.20: N = 197
arm 2 own card minus Barnum profile, dz0.30: N = 88
查看完整验算程序
# Chapter 8. p08: Sample size for the pre-registered card test. Arm 1,
# blind matching: each participant sees k anonymised cards - one built
# from their own climate cell and dominant god, k - 1 decoys built from
# other people's charts (eligible decoys: p09) - and picks the one that
# fits. Pure Barnum text gives hit rate 1/k. One-sided exact binomial test
# at alpha 0.05 (0.025 when the climate card and the dominant-god card are
# two primary endpoints); smallest N with exact power >= 0.80. Arm 2,
# Barnum control: each participant rates (1-5) their own card and a
# generic profile of the Forer type; paired difference, normal theory,
# one-sided alpha 0.025, power 0.80, for standardised effects dz.
from math import exp, lgamma, log, ceil, sqrt
from statistics import NormalDist

def tail(n, c, p):
    lp, lq = log(p), log(1 - p)
    return sum(exp(lgamma(n + 1) - lgamma(x + 1) - lgamma(n - x + 1)
                   + x * lp + (n - x) * lq) for x in range(c, n + 1))

def power(n, p0, p1, alpha):
    c = int(n * p0)
    while tail(n, c, p0) > alpha:
        c += 1
    return tail(n, c, p1), c

def n_needed(p0, p1, alpha):
    z = NormalDist().inv_cdf
    na = ((z(1 - alpha) * sqrt(p0 * (1 - p0)) + z(0.80) *
           sqrt(p1 * (1 - p1))) / (p1 - p0)) ** 2
    n = max(10, int(0.8 * na))
    while power(n, p0, p1, alpha)[0] < 0.80:
        n += 1
    return n

if __name__ == "__main__":
    print("   arm 1   k   chance   true hit   N (one end)   N (two ends)")
    for k, p1 in ((3, 0.40), (3, 0.45), (3, 0.50), (5, 0.30), (5, 0.35)):
        print("           {}   {:.4f}   {:.2f}       {:>8}      {:>8}".format(
            k, 1 / k, p1, n_needed(1 / k, p1, 0.05),
            n_needed(1 / k, p1, 0.025)))
    pw, c = power(300, 1 / 3, 1 / 3, 0.05)
    print("   N = 300, k = 3: reject if hits >= {}; false-positive rate"
          " {:.4f}".format(c, pw))
    z = NormalDist().inv_cdf
    for dz in (0.15, 0.20, 0.30):
        print("   arm 2   own card minus Barnum profile, dz = {:.2f}:"
              " N = {}".format(dz, ceil(((z(0.975) + z(0.80)) / dz) ** 2)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《子平调候张力与十神人性博弈母典》第 8 章「实践:样盘 A 的解读卡,以及一份对照巴纳姆的盲配检验」(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-119 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026119,
  author    = {Fan, Yafu},
  title     = {The Ziping Climate-Tension and Ten-Gods Codex of Human Games},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论