关于「实证闭环:保单 IRR 拆解器、预注册方案与证伪条件」,《储蓄型保单内部收益率拆解》给出的定量结论是什么?
结论:第八章给出蒸馏小工具「保单 IRR 拆解器」、个人日志、门店披露流程与预注册检验方案。书中以标注框区分已发表的证据、可运行的 Python 计算(21 段程序)、原典与作者的假说。
数据口径:书中保单、保费、现金价值、费率与死亡率均为示意,不点名、不评价任何保险公司或产品;全部计算由书中程序生成;本书属于非个人化理财建议,不构成法律或税务意见,法条与税务问题以上架前人工复核为准。 第一章显示,示意保单持有三十年的收益率只有 0.02462,现金价值仅为同样交费买国债阶梯终值的 0.9906。
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
| exit | CV/paid | IRR | CV/bond | BTID needs |
|---|---|---|---|---|
| 1 | 0.4116 | -0.58840 | 0.4016 | -0.55789 |
| 3 | 0.6587 | -0.19467 | 0.6269 | -0.16327 |
| 5 | 0.8081 | -0.07022 | 0.7499 | -0.04684 |
| 8 | 0.9589 | -0.00932 | 0.8567 | 0.00657 |
| 10 | 1.0276 | 0.00494 | 0.8948 | 0.01786 |
| 12 | 1.0951 | 0.01211 | 0.9077 | 0.02335 |
| 15 | 1.1964 | 0.01711 | 0.9208 | 0.02690 |
| 20 | 1.3869 | 0.02121 | 0.9435 | 0.02959 |
| 25 | 1.6079 | 0.02333 | 0.9668 | 0.03081 |
| 30 | 1.8639 | 0.02462 | 0.9906 | 0.03144 |
| 40 | 2.5050 | 0.02613 | 1.0400 | 0.03195 |
| first listed exit with CV >= premiums paid: 10 | ||||
| first listed exit with IRR >= bond 0.025: 40 |
查看完整验算程序
# Chapter 8. p81: The core of the Policy IRR Decomposer (distill/tool.py).
# Input: a premium paid at the start of each of the first m years and a
# guaranteed cash-value table {exit year: cash value}, here the values
# printed by p11 (rounded); a bond-ladder yield (0.025) and a flat yearly
# term-cover cost (690, close to the 30-year average of p21), both
# illustrative. For each listed exit year we compute the IRR (bisection,
# fsum), the bond pot and CV / bond, and the net return the invested
# difference must earn to match the cash value. Same formulas as tool.py.
# 模型:保单 IRR 拆解器的核心计算,与 distill/tool.py 完全一致。
import math
PREM, PAY, BOND, TERM = 10000.0, 10, 0.025, 690.0
CV = {1: 4116, 3: 19762, 5: 40403, 8: 76716, 10: 102756, 12: 109513,
15: 119640, 20: 138695, 25: 160786, 30: 186395, 40: 250499}
def bisect(f, lo, hi):
for _ in range(200):
mid = 0.5 * (lo + hi)
if f(lo) * f(mid) <= 0.0:
hi = mid
else:
lo = mid
return 0.5 * (lo + hi)
def premiums(n):
return [PREM if t < min(n, PAY) else 0.0 for t in range(n)]
def irr(n, cv):
cf = [-p for p in premiums(n)] + [cv]
return bisect(lambda r: math.fsum(c / (1.0 + r) ** t
for t, c in enumerate(cf)),
-0.99, 1.0)
def bond_pot(n, y):
return math.fsum(p * (1.0 + y) ** (n - t)
for t, p in enumerate(premiums(n)))
def btid_needed(n, cv):
def gap(r):
pot = math.fsum((p - TERM) * (1.0 + r) ** (n - t)
for t, p in enumerate(premiums(n)))
return pot - cv
return bisect(gap, -0.99, 1.0)
if __name__ == "__main__":
print("exit CV/paid IRR CV/bond BTID needs")
first_even = first_bond = None
for n in sorted(CV):
paid = PREM * min(n, PAY)
r = irr(n, CV[n])
if first_even is None and CV[n] >= paid:
first_even = n
if first_bond is None and r >= BOND:
first_bond = n
print("{:>4} {:>7.4f} {:>8.5f} {:>7.4f} {:>9.5f}"
.format(n, CV[n] / paid, r, CV[n] / bond_pot(n, BOND),
btid_needed(n, CV[n])))
print("first listed exit with CV >= premiums paid: {}".format(first_even))
print("first listed exit with IRR >= bond {:.3f}: {}"
.format(BOND, first_bond))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《储蓄型保单内部收益率拆解、哈雷生命表精算与买定期投差额边界母典》第 8 章「实证闭环:保单 IRR 拆解器、预注册方案与证伪条件」(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-220 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026220,
author = {Fan, Yafu},
title = {Decomposing the IRR of Savings Insurance: Halley's Life Table and the Limits of Buying Term and Investing the Difference},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}