关于「限量席位:k 个相同席位的第 k+1 价与标价对照」,《二价暗标与最优保留价》给出的定量结论是什么?
结论:第六章把单件推广到限量席位:申请者为席位两倍时,第 k+1 价的利润是最优标价的 1.0806 倍。
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: 1.0806
| k | n | best posted p | profit | auction r* | auction/posted | P(lot) |
|---|---|---|---|---|---|---|
| 4 | 4 | 35000 | 62500 | 62500 | 1.0000 | 0.0000 |
| 4 | 8 | 41000 | 103778 | 112139 | 1.0806 | 0.3101 |
| 4 | 12 | 45000 | 124999 | 138499 | 1.1080 | 0.4897 |
| 8 | 8 | 35000 | 125000 | 125000 | 1.0000 | 0.0000 |
| 8 | 16 | 41000 | 216963 | 231184 | 1.0655 | 0.3257 |
| 8 | 24 | 45500 | 262180 | 284802 | 1.0863 | 0.5259 |
查看完整验算程序
# Chapter 6. p62: a posted seat price versus the (k+1)-th price rule.
# Same circle as p61: seat values iid U[20,000, 60,000] yuan, cost
# c = 10,000 per seat (illustrative). Posted price p: every applicant
# with v >= p accepts; if more than k accept, seats go by public lot.
# Expected profit = (p - c) * E[min(k, Binomial(n, 1 - F(p)))], exact
# sum over the binomial with math.fsum; p searched on a 500-yuan grid.
# Auction: (k+1)-th price with reserve r* = 35,000 (p61 formula).
# Also reported: P(lot), the chance that more than k accept, so that a
# public lot may seat a lower-value applicant ahead of a higher one.
import math
L, H, C = 20000.0, 60000.0, 10000.0
W = H - L
def posted(k, n, p):
q = 1.0 - (p - L) / W
terms = [math.comb(n, j) * q ** j * (1 - q) ** (n - j) * min(k, j)
for j in range(n + 1)]
return (p - C) * math.fsum(terms)
def p_top(v, k, n):
F = (v - L) / W
return math.fsum(math.comb(n - 1, j) * (1 - F) ** j
* F ** (n - 1 - j) for j in range(min(k, n)))
def auction(k, n, r=35000.0, m=2000):
h = (H - r) / m
t = []
for i in range(m + 1):
v = r + i * h
w = 1 if i in (0, m) else (4 if i % 2 else 2)
t.append(w * (2.0 * v - H - C) * p_top(v, k, n) / W)
return n * h / 3.0 * math.fsum(t)
def over(k, n, p):
q = 1.0 - (p - L) / W
return math.fsum(math.comb(n, j) * q ** j * (1 - q) ** (n - j)
for j in range(k + 1, n + 1))
print("k n best posted p profit auction r* auction/posted"
" P(lot)")
for k in (4, 8):
for n in (k, 2 * k, 3 * k):
grid = [20000.0 + 500.0 * i for i in range(81)]
p = max(grid, key=lambda x: (posted(k, n, x), -x))
a = auction(k, n)
print("%d %-2d %9.0f %7.0f %8.0f %.4f %.4f"
% (k, n, p, posted(k, n, p), a, a / posted(k, n, p),
over(k, n, p)))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 6 章「限量席位:k 个相同席位的第 k+1 价与标价对照」(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}
}