关于「门槛博弈的均衡清单:退款承诺到底改变了什么」,《保证合约与预售成团》给出的定量结论是什么?
结论:第一章精确枚举五人门槛博弈:能开工的均衡所占比例在不退规则下为 0.909,在退款规则下为 0.625;在对称混合均衡中,退款清除了开工概率为 0.2161 的低均衡,剩余均衡的开工概率为 0.8256。
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: 0.625, 0.909
| N | 5 K = 3 V = 10 C = 4 profiles = 32 |
| rule pledgers NE count batch made | |
| keep | 0 1 no |
| keep | 3 10 yes |
| refund | 0 1 no |
| refund | 1 5 no |
| refund | 3 10 yes |
| keep equilibria 11, batch made in 10, share 0.909 | |
| refund equilibria 16, batch made in 10, share 0.625 | |
| refund adds equilibria with 1 pledger: refunded pledgers | |
| lose nothing, so a lone pledger has no reason to withdraw |
查看完整验算程序
# Chapter 1. p11: every pure-strategy Nash equilibrium of a small
# threshold game, with and without a refund when the threshold fails.
# Model (Palfrey and Rosenthal 1984 type): N buyers each decide to
# pledge (cost C) or not; the batch is made iff at least K pledge, and
# then every buyer, pledger or not, gets value V (the batch exists and
# later buyers can free-ride on it). Rule "keep": a failed pledge is
# lost. Rule "refund": a failed pledge is returned in full.
# Parameters are illustrative: N = 5, K = 3, V = 10, C = 4.
# Exact enumeration of all 2 ** N profiles; standard library only.
import itertools
N, K, V, C = 5, 3, 10, 4
def payoff(mine, others, rule):
made = mine + others >= K
if mine:
if made:
return V - C
return 0 if rule == "refund" else -C
return V if made else 0
def is_nash(prof, rule):
k = len([x for x in prof if x])
for x in prof:
others = k - x
now = payoff(x, others, rule)
dev = payoff(1 - x, others, rule)
if dev > now:
return False
return True
def table(rule):
found = {}
for prof in itertools.product((0, 1), repeat=N):
if is_nash(prof, rule):
k = len([x for x in prof if x])
found[k] = found.get(k, 0) + 1
return found
if __name__ == "__main__":
print("N = {} K = {} V = {} C = {} profiles = {}".format(
N, K, V, C, 2 ** N))
print("rule pledgers NE count batch made")
tot = {}
for rule in ("keep", "refund"):
f = table(rule)
for k in sorted(f):
print("{:7s} {:8d} {:9d} {}".format(
rule, k, f[k], "yes" if k >= K else "no"))
good = len([1 for k in f for _ in range(f[k]) if k >= K])
allne = len([1 for k in f for _ in range(f[k])])
tot[rule] = (good, allne)
for rule in ("keep", "refund"):
g, a = tot[rule]
print("{:7s} equilibria {:2d}, batch made in {:2d}, share {:.3f}"
.format(rule, a, g, g / a))
print("refund adds equilibria with 1 pledger: refunded pledgers")
print("lose nothing, so a lone pledger has no reason to withdraw")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 1 章「门槛博弈的均衡清单:退款承诺到底改变了什么」(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}
}