关于「东西方原典考辨:米芾《画史》与瓦萨里《名人传》」,《收藏品回报去偏》给出的定量结论是什么?

结论:教皇像一表:复制品通过率为 0.10 时,风格判断的似然比是 9.50,通过后的原作概率是 0.905;复制水平升到 0.60,似然比只剩 1.58,后验 0.613;到了 0.90,也就是朱利奥·罗马诺遇到的情形,似然比 1.06,后验 0.514,几乎等于没有看。相反,知情人的记号似然比是 95,后验 0.990。如果风格判断通过之后,再发现画上没有那个记号,原作的概率就从 0.514 跌到 0.0506。瓦萨里那一指,比朱利奥的一双眼睛有用得多。

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。 与结论吻合的打印值: 0.0506, 0.10, 0.514, 0.60, 0.613, 0.90

Cupid: paid 200, maker got 30, markup 6.67 times
maker share 0.150, middleman share 0.850 (1 crown1 ducat)
copy pass rate LR P(original|pass)
0.109.50 0.905
0.303.17 0.760
0.601.58 0.613
0.801.19 0.543
0.901.06 0.514
insider's mark: LR 95, P(original|mark) 0.990
eye check, then mark absent: P(original) 0.0506
查看完整验算程序
# Chapter 7. p72: two stories from Vasari read as numbers
# 1) The sleeping Cupid (Life of Michelangelo): the cardinal paid 200
#    ducats; Michelangelo was sent 30 crowns. Treating one crown as one
#    ducat is an illustrative assumption, so the shares are rough.
# 2) Andrea del Sarto's copy of Raphael's Leo X (Life of Andrea del Sarto)
#    deceived Giulio Romano, who had worked on the original. Model: an eye
#    check passes a genuine work with probability 0.95 and a copy with
#    probability FALSE, rising with copy quality; a private mark known to
#    one insider passes a genuine work with 0.95 and a copy with 0.01.
#    Prior 0.5 that the panel shown is the original. Rates illustrative.
PAID, MAKER = 200.0, 30.0
SENS = 0.95
FALSES = [0.10, 0.30, 0.60, 0.80, 0.90]


def post(prior, se, fa):
    o = prior / (1.0 - prior) * se / fa
    return o / (1.0 + o)


def main():
    print("Cupid: paid {:.0f}, maker got {:.0f}, markup {:.2f} times".format(
        PAID, MAKER, PAID / MAKER))
    print("maker share {:.3f}, middleman share {:.3f} (1 crown = 1 ducat)"
          .format(MAKER / PAID, 1.0 - MAKER / PAID))
    print("{:>16}{:>8}{:>18}".format("copy pass rate", "LR",
                                    "P(original|pass)"))
    for fa in FALSES:
        print("{:>16.2f}{:>8.2f}{:>18.3f}".format(fa, SENS / fa,
                                                post(0.5, SENS, fa)))
    print("insider's mark: LR {:.0f}, P(original|mark) {:.3f}".format(
        SENS / 0.01, post(0.5, SENS, 0.01)))
    print("eye check, then mark absent: P(original) {:.4f}".format(
        post(post(0.5, SENS, 0.90), 1.0 - SENS, 0.99)))


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 7 章「东西方原典考辨:米芾《画史》与瓦萨里《名人传》」(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