关于「退改期权阶梯:用再售概率给每一级定价、服务逃逸与淡旺季动态阶梯」,《摄影场馆的装箱、退改与折旧》给出的定量结论是什么?
结论:第三章用再售概率给退改阶梯每一级定价,阶梯下每单净收入为 1.1035,高于不退款的 1.0261 与全额退款的 1.0113。
数据口径:书中参数为示意;全部计算由书中程序生成;本书不是个人化的税务、法律或投资建议。 书中以标注框区分已发表的证据、可运行的 Python 计算(22 段程序)、原典与作者的假说。
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: 1.0113, 1.0261, 1.1035
| policy | holds | E[refund] | E[resale] | net psi=0 | net psi=0.15 |
| no refund | 0.000 | 0.0000 | 0.0261 | 1.0261 | 1.0261 |
| ladder p31 | 0.015 | 0.0849 | 0.1042 | 1.1035 | 1.1162 |
| full refund | 0.150 | 0.2600 | 0.1013 | 1.0113 | 1.0368 |
| ladder: a genuine canceller gets back 0.4274 of the price | |||||
| on average; resale income rises from 0.0261 to 0.1042 |
查看完整验算程序
# Chapter 3. p32: the refund ladder as an option sold to the client.
# A client books 30 days ahead. With probability 0.20 her plan breaks;
# the day she learns it is uniform on 0 to 30 days before the shoot.
# She tells the venue at once only if telling pays: always under a
# refund, with probability 0.25 under a strict no-refund rule. A told
# slot is resold with q(d) from p31 at 0.9 of the price. Genuine
# clients accept a price of base + (1 + psi) E[refund]. Generous late
# refunds also attract holds: a share 0.15 x r(1 day) of bookings are
# speculative, and 0.6 of them cancel in the last 3 days.
# Per 1 yuan of base price. Illustrative parameters.
import math
KAPPA, LEAD, DELTA, TELL0, L, MEAN = 0.20, 30, 0.9, 0.25, 1.5, 12.0
HOLD, HCANCEL, HLATE = 0.15, 0.6, 3
def q(d):
return 1 - math.exp(-L * (1 - math.exp(-d / MEAN)))
def ladder(d):
for lo, r in ((30, 0.65), (14, 0.55), (7, 0.40), (3, 0.25),
(1, 0.10)):
if d >= lo:
return r
return 0.0
def avg(f, top):
grid = [k / 10 for k in range(0, top * 10 + 1)]
return math.fsum(f(d) for d in grid) / len(grid)
def policy(refund, tell):
h = HOLD * refund(1)
eg = (1 - h) * KAPPA * avg(refund, LEAD)
eh = h * HCANCEL * avg(refund, HLATE)
sg = (1 - h) * KAPPA * tell * DELTA * avg(q, LEAD)
sh = h * HCANCEL * DELTA * avg(q, HLATE)
return h, eg, eh, sg + sh
rows = [("no refund", lambda d: 0.0, TELL0),
("ladder p31", ladder, 1.0),
("full refund", lambda d: 1.0, 1.0)]
print("policy holds E[refund] E[resale] net psi=0 net psi=0.15")
for name, f, t in rows:
h, eg, eh, es = policy(f, t)
n0 = 1 - eh + es
n1 = 1 + 0.15 * eg - eh + es
print("{:11s} {:5.3f} {:9.4f} {:9.4f} {:9.4f} {:12.4f}".format(
name, h, eg + eh, es, n0, n1))
h, eg, eh, es = policy(ladder, 1.0)
print("ladder: a genuine canceller gets back {:.4f} of the price".format(
avg(ladder, LEAD)))
print("on average; resale income rises from {:.4f} to {:.4f}".format(
policy(lambda d: 0.0, TELL0)[3], es))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《多规格实景空间时段一维装箱、动态退改期权阶梯与高损耗器材加速折旧运筹》第 3 章「退改期权阶梯:用再售概率给每一级定价、服务逃逸与淡旺季动态阶梯」(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-173 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026173,
author = {Fan, Yafu},
title = {One-Dimensional Time-Slot Packing for Multi-Format Studio Sets, Dynamic Cancellation-Option Ladders and Accelerated Depreciation of High-Wear Equipment},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}