关于「「我自己会安排好」的高估幅度:规划谬误与健身房合同」,《截止期与时间动机》给出的定量结论是什么?

结论:第五章用毕业论文与健身房合同两组数据求出几乎相同的高估幅度 1.637 与 1.628,要有八成把握按期完成需要把自己的估计乘以 2.690。

数据口径:书中参数为示意;全部计算由书中程序生成;本书不是个人化的经营、法律或理财建议。 书中以标注框区分已发表的证据、可运行的 Python 计算(18 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: 1.637, 2.690

ratio of means actual / best1.637
lognormal fit of R: mu 0.384, sigma 0.720
median R 1.468, mean R 1.902
check: P(R <1) 0.297, P(R <= 1.434) 0.487
target on-time multiplier days for a 33.9-day guess
0.501.468 49.8
0.702.141 72.6
0.802.690 91.2
0.903.692 125.1
0.954.795 162.5
查看完整验算程序
# Chapter 5. p51: How large a buffer the planning fallacy demands. Source:
# Buehler, Griffin and Ross (1994), Study 1, honours theses, 33 analysed:
# best guess 33.9 days, actual 55.5; 29.7 % done by the best guess;
# pessimistic guess 48.6 days, 48.7 % done by it. We fit a lognormal to
# R = actual / best guess from the two shares (the pessimistic guess is
# 48.6 / 33.9 of the best guess on average; an approximation) and ask
# what multiplier on one's own guess gives 50 to 95 % on-time.
import math

def phi(z):
    return 0.5 * (1.0 + math.erf(z / math.sqrt(2.0)))

def probit(p):
    lo, hi = -8.0, 8.0
    for _ in range(200):
        mid = (lo + hi) / 2.0
        if phi(mid) < p:
            lo = mid
        else:
            hi = mid
    return (lo + hi) / 2.0

BEST, ACTUAL, PESS = 33.9, 55.5, 48.6
P_BEST, P_PESS = 0.297, 0.487

if __name__ == "__main__":
    z1, z2 = probit(P_BEST), probit(P_PESS)
    x2 = math.log(PESS / BEST)
    sig = x2 / (z2 - z1)
    mu = -z1 * sig
    print("ratio of means actual / best: %.3f" % (ACTUAL / BEST))
    print("lognormal fit of R: mu %.3f, sigma %.3f" % (mu, sig))
    print("median R %.3f, mean R %.3f" % (math.exp(mu),
                                         math.exp(mu + sig * sig / 2)))
    print("check: P(R <= 1) %.3f, P(R <= %.3f) %.3f"
          % (phi(-mu / sig), PESS / BEST, phi((x2 - mu) / sig)))
    print("target on-time  multiplier  days for a 33.9-day guess")
    for p in (0.5, 0.7, 0.8, 0.9, 0.95):
        m = math.exp(mu + probit(p) * sig)
        print("     %.2f         %6.3f       %6.1f" % (p, m, m * BEST))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《双曲贴现偏好反转、截止期承诺装置与时间动机方程母典》第 5 章「「我自己会安排好」的高估幅度:规划谬误与健身房合同」(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-199 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026199,
  author    = {Fan, Yafu},
  title     = {Hyperbolic-Discounting Preference Reversals, Deadline Commitment Devices and the Temporal Motivation Equation},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论