关于「真实进度展示规则:计数、截止、余量与退款的逐条审计」,《保证合约与预售成团》给出的定量结论是什么?
结论:这张表有两点值得注意。第一,线性外推在开头高估、在后段低估。第 3 天按计划进度有 42 单,线性外推得 196.0 单,远高于真实预期;第 12 天按计划进度有 84 单,线性外推只得 98.0 单,似乎凑不够,可是按剩余的预期 31.1 单计算,达标概率为 0.9989。第二,真实计数一旦落后计划,达标概率下降得很快:第 10 天若只有计划的八成,即 60 单,达标概率为 0.5464;第 12 天若只有计划的八成,即 67 单,只剩 0.3888。
数据口径:书中参数为示意;全部计算由书中程序生成。
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: 0.3888, 0.5464, 0.9989, 10, 12, 196.0
| K = 100 | expected total = 115; P(made) at launch 0.9284 | ||||
| day | m vs plan | m | straight line | rest mean | P(made | m) |
| 3 | 0.8 | 34 | 158.7 | 73.0 | 0.8101 |
| 3 | 1.0 | 42 | 196.0 | 73.0 | 0.9692 |
| 3 | 1.2 | 50 | 233.3 | 73.0 | 0.9982 |
| 7 | 0.8 | 48 | 96.0 | 54.4 | 0.6454 |
| 7 | 1.0 | 61 | 122.0 | 54.4 | 0.9878 |
| 7 | 1.2 | 73 | 146.0 | 54.4 | 1.0000 |
| 10 | 0.8 | 60 | 84.0 | 40.4 | 0.5464 |
| 10 | 1.0 | 75 | 105.0 | 40.4 | 0.9962 |
| 10 | 1.2 | 90 | 126.0 | 40.4 | 1.0000 |
| 12 | 0.8 | 67 | 78.2 | 31.1 | 0.3888 |
| 12 | 1.0 | 84 | 98.0 | 31.1 | 0.9989 |
| 12 | 1.2 | 101 | 117.8 | 31.1 | 1.0000 |
查看完整验算程序
# Chapter 6. p61: what a true progress count says. With the 14-day
# arrival profile of p32 (expected total 115, K = 100), a page that
# shows the real count m after day t lets anyone compute the honest
# chance of success P(N_rest >= K - m), N_rest ~ Poisson(rest of the
# expected total). A straight-line projection m x 14 / t ignores the
# closing burst and reads the quiet middle as failure. Rows: m equal
# to 0.8, 1.0 and 1.2 times the count expected by day t.
import math
K, MU = 100, 115.0
W = [3.0, 1.8] + [0.6] * 10 + [1.4, 2.6]
def pois_tail(k, mu):
if k <= 0:
return 1.0
terms = [math.exp(j * math.log(mu) - mu - math.lgamma(j + 1))
for j in range(k)]
return 1.0 - math.fsum(terms)
if __name__ == "__main__":
wsum = math.fsum(W)
rates = [MU * w / wsum for w in W]
print("K = {} expected total = {:.0f}; P(made) at launch {:.4f}"
.format(K, MU, pois_tail(K, MU)))
print("day m vs plan m straight line rest mean P(made | m)")
for t in (3, 7, 10, 12):
done = math.fsum(rates[:t])
rest = math.fsum(rates[t:])
for f in (0.8, 1.0, 1.2):
m = int(round(f * done))
line = m * len(W) / t
print("{:3d} {:9.1f} {:5d} {:14.1f} {:10.1f} {:12.4f}".format(
t, f, m, line, rest, pois_tail(K - m, rest)))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《保证合约预售成团阈值、退款承诺与公共品搭便车均衡母典》第 6 章「真实进度展示规则:计数、截止、余量与退款的逐条审计」(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-316 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026316,
author = {Fan, Yafu},
title = {Assurance Contracts and Pre-Sale Group Thresholds: Refund Promises and Free-Riding Equilibria in Public Goods},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}