关于「实证闭环:配置上限、尽调清单与证伪方案」,《收藏品回报去偏》给出的定量结论是什么?

结论:第八章演示高净值客户另类资产配置上限:宣传口径的模型配置为 0.251,扣除各项成本后为零;并给出藏品入手前的尽调清单、蒸馏小工具「收藏回报去偏器」与一份预注册证伪方案。书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。

数据口径:书中参数为示意;全部计算由书中程序生成;本书非个人化法律或理财建议,费率与法规以上架前人工复核为准。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。

English version

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: 0.251, 20

RF 0.04, risk aversion 3, piece sd 0.25
casenetweightdividend .05
headline, no costs0.08700.2510.0000
debiased, H 5, L 0.10-0.02640.0000.0758
debiased, H 10, L 0.100.00400.0000.0454
debiased, H 20, L 0.050.02840.0000.0210
headline, H 10, L 0.050.03690.0000.0125
查看完整验算程序
# Chapter 8. p81: an allocation cap for alternative assets, step by step
# Model (illustrative parameters; growth figures from Korteweg et al.):
#   net = G - fee drag - upkeep - forgery drag - forced-sale drag
#   fee drag      = 1 - (1 - C)**(1/H) for a round trip C over H years
#   forgery drag  = residual fake chance F times (1 - K), spread over H
#   forced-sale   = yearly chance L of a cash need times the deadline
#                   discount D (0.1781 = 1 - 0.8219, the 3-month row of
#                   p62), counted once per holding period
# Merton-style weight w = (net - RF) / (GAMMA * S*S), floored at zero.
# The last column is the yearly "enjoyment dividend" a holder must value
# the piece at for a 0.05 weight to make sense.
import math

C, UPKEEP, F, K = 0.25, 0.01, 0.03, 0.05
D = 1.0 - 0.8219
RF, GAMMA, S = 0.04, 3.0, 0.25
CASES = [("headline, no costs", 0.087, None, 0.0),
         ("debiased, H 5, L 0.10", 0.063, 5, 0.10),
         ("debiased, H 10, L 0.10", 0.063, 10, 0.10),
         ("debiased, H 20, L 0.05", 0.063, 20, 0.05),
         ("headline, H 10, L 0.05", 0.087, 10, 0.05)]


def net_return(g, h, lam):
    if h is None:
        return g
    fee = 1.0 - (1.0 - C) ** (1.0 / h)
    fake = F * (1.0 - K) / h
    forced = min(1.0, lam * h) * D / h
    return math.fsum([g, -fee, -UPKEEP, -fake, -forced])


def main():
    print("RF {:.2f}, risk aversion {:.0f}, piece sd {:.2f}".format(
        RF, GAMMA, S))
    print("{:<24}{:>8}{:>8}{:>14}".format("case", "net", "weight",
                                       "dividend .05"))
    for name, g, h, lam in CASES:
        r = net_return(g, h, lam)
        w = max(0.0, (r - RF) / (GAMMA * S * S))
        need = max(0.0, 0.05 * GAMMA * S * S + RF - r)
        print("{:<24}{:>8.4f}{:>8.3f}{:>14.4f}".format(name, r, w, need))


if __name__ == "__main__":
    main()

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《收藏品重复成交指数选择偏差、赝品先验与另类资产流动性折价母典》第 8 章「实证闭环:配置上限、尽调清单与证伪方案」(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-208 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026208,
  author    = {Fan, Yafu},
  title     = {Repeat-Sales Selection Bias in Collectibles, Forgery Priors and the Illiquidity Discount of Alternative Assets},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论