关于「说真话的暗标:占优策略、过度出价、托价与承诺揭示」,《二价暗标与最优保留价》给出的定量结论是什么?
结论:第四章讲二价暗标的说真话性质、过度出价与托价,暗中托价每笔从赢家身上拿走 16,000 元,并用 SHA-256 承诺与揭示锁定出价与保留价。
数据口径:书中的估值分布、保有价值、买家人数与价格均为示意;全部计算由书中程序生成;本书是非个人化经营建议,也不是法律意见,正式拍卖以上架前人工复核的当地法规为准;紧迫感只能来自真实库存与真实截止,不虚构竞价者,不托价。 第七章考辨《宋史》的买扑与苏辙「立定中价,不许添刬」之议(示意模型中按自身估计出价的违约概率达 0.9691)以及希罗多德记载的巴比伦婚配拍卖(收入是嫁资的 3.373 倍)。
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: 16000
| A. overbidding factor, mean price, P(regret), mean regret | |
| d | 0.00 price 127926 P(regret) 0.0000 regret 0 |
| d | 0.05 price 134322 P(regret) 0.3018 regret 3698 |
| d | 0.10 price 140719 P(regret) 0.5110 regret 7735 |
| d | 0.15 price 147115 P(regret) 0.6600 regret 12035 |
| B. honest second-price mean 128000, with a hidden shill 144000 | |
| transfer from the winning client per sale | 16000 yuan |
| winner's expected surplus: honest 16000, shill 0 | |
| if shills are expected: bids fall to first-price, mean 128000 |
查看完整验算程序
# Chapter 4. p42: two ways a second-price sealed bid goes wrong.
# n = 4 bidders, values iid U[80,000, 160,000] yuan (illustrative).
# A. Bidders overbid by a factor (1 + d), as laboratory subjects often
# do in second-price auctions; seeded Monte Carlo, 20,000 auctions.
# A winner "regrets" when the price paid exceeds own value.
# B. A dishonest house reads the sealed bids and inserts a fake bid
# just under the top one (shill bid): the price becomes the top bid.
# Exact: honest E[v_(2)] = L + W (n-1)/(n+1); shill E[v_(1)] =
# L + W n/(n+1). If bidders expect it, they bid as in first-price
# and the house is back to the honest mean: the gain exists only
# while the deception is hidden, which is why it is fraud.
import math
import random
from fractions import Fraction as Fr
L, H, N, RUNS = 80000, 160000, 4, 20000
rng = random.Random(20261113)
draws = [sorted((rng.uniform(L, H) for _ in range(N)), reverse=True)
for _ in range(RUNS)]
print("A. overbidding factor, mean price, P(regret), mean regret")
for d in (0.0, 0.05, 0.10, 0.15):
prices, regrets = [], []
for v in draws:
p = min(v[1] * (1 + d), v[0] * (1 + d))
prices.append(p)
regrets.append(max(0.0, p - v[0]))
hit = len([x for x in regrets if x > 0]) / RUNS
pos = [x for x in regrets if x > 0]
mr = math.fsum(pos) / len(pos) if pos else 0.0
print(" d = %.2f price %7.0f P(regret) %.4f regret %6.0f"
% (d, math.fsum(prices) / RUNS, hit, mr))
W = H - L
honest = L + W * Fr(N - 1, N + 1)
shill = L + W * Fr(N, N + 1)
surplus = W * Fr(1, N + 1)
print("B. honest second-price mean %d, with a hidden shill %d"
% (honest, shill))
print(" transfer from the winning client per sale: %d yuan" % surplus)
print(" winner's expected surplus: honest %d, shill 0" % surplus)
print(" if shills are expected: bids fall to first-price, mean %d"
% (L + Fr(N - 1, N) * (W * Fr(N, N + 1))))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《维克里二价暗标、迈尔森最优保留价与孤品奢包限量席位激励相容拍卖母典》第 4 章「说真话的暗标:占优策略、过度出价、托价与承诺揭示」(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-237 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026237,
author = {Fan, Yafu},
title = {Vickrey Second-Price Sealed Bids and Myerson Optimal Reserve Prices: Incentive-Compatible Auctions for One-of-a-Kind Luxury Pieces and Limited Seats},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}