关于「托管、自动执行与分级制裁」,《时间不一致与反掀桌质押》给出的定量结论是什么?
结论:第六章比较托管里程碑与先付后付,并说明监察有误时一次性除名会误除 0.506 的守约者,分级制裁合规率仍有 0.963。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是法律意见,也不是个人化财务或经营建议,所有例子都是合成情境,价格一律为无币种示意单位。 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。
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: 0.506, 0.963
| regime | FP | compliance | steady_expelled | members_left |
|---|---|---|---|---|
| heavy | 0.03 | 1.000 | 0.506 | 97 |
| graduated | 0.03 | 0.963 | 0.000 | 181 |
| verified | 0.003 | 1.000 | 0.087 | 184 |
| closed form, steady member expelled under heavy: | ||||
| FP = 0.03: 1 - (1 - FP)^24 = 0.519 | ||||
| FP = 0.003: 1 - (1 - FP)^24 = 0.070 |
查看完整验算程序
# Chapter 6. p62: graduated sanctions versus one heavy sanction
# 200 members over 24 rounds; 160 are steady compliers, 40 are
# opportunists who breach when the gain g ~ U(0, 10) exceeds the
# expected sanction DET * (the sanction their next flag would bring).
# A real breach is flagged with probability DET; a compliant act is
# flagged by mistake with probability FP. Losing membership costs 40.
# heavy first flag -> expulsion
# graduated flags 1, 2, 3 -> fines 3, 6, 12; flag 4 -> expulsion
# verified heavy, but each flag is checked by a third party first,
# which cuts FP tenfold
# Seeded simulation; values illustrative and unit-free.
import random
N_STEADY, N_OPP, ROUNDS, DET, V = 160, 40, 24, 0.6, 40.0
LADDER = [3.0, 6.0, 12.0, V]
def run(regime, fp, seed_rng):
rng = seed_rng
flags = [0] * (N_STEADY + N_OPP)
out = [False] * (N_STEADY + N_OPP)
acts, ok = 0, 0
for _ in range(ROUNDS):
for i in range(N_STEADY + N_OPP):
if out[i]:
continue
if regime == "graduated":
nxt = LADDER[min(flags[i], 3)]
else:
nxt = V
breach = i >= N_STEADY and rng.uniform(0, 10) > DET * nxt
acts += 1
ok += 0 if breach else 1
hit = rng.random() < (DET if breach else fp)
if hit:
flags[i] += 1
if regime != "graduated" or flags[i] >= 4:
out[i] = True
lost = 0
for i in range(N_STEADY):
lost += 1 if out[i] else 0
stay = 0
for i in range(N_STEADY + N_OPP):
stay += 0 if out[i] else 1
return ok / acts, lost / N_STEADY, stay
print("regime FP compliance steady_expelled members_left")
for regime, fp in (("heavy", 0.03), ("graduated", 0.03),
("verified", 0.003)):
c, l, s = run(regime, fp, random.Random(41362))
print("{:<10} {:<6} {:>10.3f} {:>15.3f} {:>12}".format(
regime, fp, c, l, s))
print("closed form, steady member expelled under heavy:")
for fp in (0.03, 0.003):
print(" FP = {}: 1 - (1 - FP)^24 = {:.3f}".format(
fp, 1 - (1 - fp) ** ROUNDS))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《动态时间不一致博弈、收益私有化亏损转嫁与反掀桌质押契约母典》第 6 章「托管、自动执行与分级制裁」(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-422 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026422,
author = {Fan, Yafu},
title = {Dynamic Time-Inconsistency, Asymmetric Freeroll Reneging and Collateralised Anti-Hold-Up Contracts},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}