关于「享乐适应的指数基线:半衰期的估计与可识别性」,《收入幸福饱和与奢侈折旧》给出的定量结论是什么?

结论:第四章建立享乐适应的指数基线,并给出由第 0、30、90 天三次回访解出半衰期的封闭公式及其适用范围。

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建议,也不是理财或投资建议。

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: 30, 90

arm true T fitted T fitted s0 fitted s_inf rmse
material21.0 20.0 8.68 6.90 0.054
experience75.0 145.0 8.15 6.59 0.102
share of the initial excess left (true parameters):
material day 300.371 day 90: 0.051
experience day 300.758 day 90: 0.435
查看完整验算程序
# Chapter 4. p41: The exponential baseline of hedonic adaptation.
# 模型:满意度 s(t) = s_inf + (s0 - s_inf) * 2 ** (-t / T),T 为半衰期(天)。
# 两类购买(示意参数,不是文献估计):实物奢品 T = 21 天,
# 体验消费 T = 75 天。每类 120 名顾客,在第 1、7、14、30、60、90、180 天
# 打 0-10 分;顾客有个人偏移(SD 1.0)与测量噪声(SD 0.8)。
# 对每个候选 T,(s0, s_inf) 是线性最小二乘;在 T 网格上取残差最小者。
import math
import random

DAYS = (1, 7, 14, 30, 60, 90, 180)
ARMS = {"material": (8.6, 6.8, 21.0), "experience": (8.4, 7.2, 75.0)}
N = 120
rng = random.Random(20261112)


def curve(t, s0, sinf, half):
    return sinf + (s0 - sinf) * 2 ** (-t / half)


def simulate(s0, sinf, half):
    rows = []
    for _ in range(N):
        off = rng.gauss(0, 1.0)
        rows.append([min(10.0, max(0.0, curve(t, s0, sinf, half) + off
                                   + rng.gauss(0, 0.8))) for t in DAYS])
    return rows


def fit(means, grid=None):
    grid = grid or [h / 2 for h in range(4, 801)]
    best = None
    for half in grid:
        e = [2 ** (-t / half) for t in DAYS]
        me, my = math.fsum(e) / len(e), math.fsum(means) / len(means)
        sxx = math.fsum((a - me) ** 2 for a in e)
        sxy = math.fsum((a - me) * (b - my) for a, b in zip(e, means))
        amp = sxy / sxx
        sinf = my - amp * me
        sse = math.fsum((sinf + amp * a - b) ** 2 for a, b in zip(e, means))
        if best is None or sse < best[0]:
            best = (sse, sinf + amp, sinf, half)
    return best


if __name__ == "__main__":
    print("arm          true T  fitted T  fitted s0  fitted s_inf  rmse")
    for arm, (s0, sinf, half) in ARMS.items():
        rows = simulate(s0, sinf, half)
        means = [math.fsum(r[j] for r in rows) / N for j in range(len(DAYS))]
        sse, f0, finf, fh = fit(means)
        print("{:11s} {:7.1f} {:9.1f} {:10.2f} {:13.2f} {:6.3f}".format(
            arm, half, fh, f0, finf, math.sqrt(sse / len(DAYS))))
    print("share of the initial excess left (true parameters):")
    for arm, (s0, sinf, half) in ARMS.items():
        print("  {:11s} day 30: {:.3f}   day 90: {:.3f}".format(
            arm, 2 ** (-30 / half), 2 ** (-90 / half)))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《收入幸福曲线饱和点争议、享乐适应速率与奢侈消费体验折旧母典》第 4 章「享乐适应的指数基线:半衰期的估计与可识别性」(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-198 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026198,
  author    = {Fan, Yafu},
  title     = {The Income-Wellbeing Satiation Debate, Hedonic Adaptation Rates and the Experiential Depreciation of Luxury Consumption},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论