关于「实践与证伪:主事者清单、试算器说明与预注册方案」,《达人经纪的幂律、抽查与买断》给出的定量结论是什么?

结论:第八章给出主事者清单、文案检查规则与预注册方案。书中以标注框区分已发表的证据、可运行的 Python 计算(22 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;本书不是个人化的法律或理财建议。 第六章把阶梯分成、买断金与抽查率合成三联定价试算器,买断金取剩余佣金的 1.712 倍。

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

H1 cohort head rate 0.0140 +/- 30%3007 signings in the cohort
expected heads among them42.1
H2 diversion 0.15 against 0.05141 creators per arm
H3 tail index 1.5, SE 0.10225 tail creators, roster 1185
H4 walk-away 0.2245 against 0.0560 creators per arm
power check, H2 at 150 per arm0.833
查看完整验算程序
# Chapter 8. p81: sample sizes for the pre-registered tests.
# H1: estimate a cohort head rate of 0.0140 (p11) to +/- 30% at 95%.
# H2: audit field test, two arms (audit rate 0.10 against p*); the
# diversion rate found at the final reconciliation is predicted to fall
# from 0.15 to 0.05.  H3: tail index 1.5 with standard error 0.10,
# SE = alpha / sqrt(n_tail); a tail share of 0.19 (p21) converts the
# tail count into a roster size.  H4: walk-away rate, flat split
# against ladder, predicted 0.2245 (449 of 2000 in p42) against 0.05.
# Two-sided alpha 0.05, power 0.80, normal approximations.
import math
from statistics import NormalDist

Z = NormalDist()
ZA, ZB = Z.inv_cdf(0.975), Z.inv_cdf(0.80)


def two_prop(p1, p2):
    pb = (p1 + p2) / 2
    a = ZA * math.sqrt(2 * pb * (1 - pb))
    b = ZB * math.sqrt(p1 * (1 - p1) + p2 * (1 - p2))
    return math.ceil(((a + b) / (p1 - p2)) ** 2)


h = 0.0140
n1 = math.ceil(ZA ** 2 * (1 - h) / (h * 0.30 ** 2))
print("H1 cohort head rate {:.4f} +/- 30%: {} signings in the cohort"
      .format(h, n1))
print("   expected heads among them: {:.1f}".format(n1 * h))
print("H2 diversion 0.15 against 0.05: {} creators per arm".format(
    two_prop(0.15, 0.05)))
n3 = math.ceil((1.5 / 0.10) ** 2)
print("H3 tail index 1.5, SE 0.10: {} tail creators, roster {}".format(
    n3, math.ceil(n3 / 0.19)))
print("H4 walk-away 0.2245 against 0.05: {} creators per arm".format(
    two_prop(0.2245, 0.05)))
print("power check, H2 at 150 per arm: {:.3f}".format(
    Z.cdf(abs(0.10) / math.sqrt(0.15 * 0.85 / 150 + 0.05 * 0.95 / 150)
          - ZA)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《网红经纪机构孵化幸存者幂律、达人飞单抽样审计博弈与违约买断期权定价》第 8 章「实践与证伪:主事者清单、试算器说明与预注册方案」(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-172 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026172,
  author    = {Fan, Yafu},
  title     = {Survivor Power Laws in Influencer-Agency Incubation, Sampling-Audit Games against Off-Platform Deals and Option Pricing of Contract Buyouts},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论