关于「同一笔预算:顶流的方差与中腰部组合的方差公式」,《达人矩阵的组合方差》给出的定量结论是什么?
结论:第一章写出同额预算的方差公式:二十名中腰部、相关 0.20 时,组合方差只有顶流的 0.1067,但地板是平均协方差。
数据口径:书中借用组合理论的方差、相关、夏普比率与均值—方差前沿,只借其算术,讨论的是营销预算的分配,不是证券投资;本书不构成任何个人化的理财或投资建议,不推荐任何金融产品。 书中的达人、回报、相关、成本与预算均为示意;全部计算由书中程序生成;所有合作都须如实标注广告,不讨论购买流量或虚假评价。
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: 0.1067, 0.20
| budget 600000 yuan; head creator: mean 0.30, sd 1.20 per yuan | |
| sd of an equal-weight portfolio of N mid-tier creators | |
| N rho | 0.00 rho=0.10 rho=0.20 rho=0.40 |
| 1 | 0.8000 0.8000 0.8000 0.8000 |
| 2 | 0.5657 0.5933 0.6197 0.6693 |
| 5 | 0.3578 0.4233 0.4800 0.5769 |
| 10 | 0.2530 0.3487 0.4233 0.5426 |
| 20 | 0.1789 0.3046 0.3919 0.5246 |
| 50 | 0.1131 0.2748 0.3718 0.5135 |
| 100 | 0.0800 0.2641 0.3649 0.5097 |
| limit | 0.0000 0.2530 0.3578 0.5060 |
| N needed to come within 10% of the limit variance | 90 40 15 |
| N | 20, rho=0.20: variance 0.1536, ratio to head variance 0.1067 |
| sd in yuan: head 720000, N | 20 portfolio 235151 |
| expected net return in yuan: head 180000, N | 20 portfolio 144000 |
| diversifiable share of one creator's variance at rho | 0.20: 0.80 |
核心公式
\[\frac{\sigma^2}{N} + \left(1 - \frac{1}{N}\right)\rho\,\sigma^2 .\]
查看完整验算程序
# Chapter 1. p11: one head creator against N mid-tier creators, same budget.
# Model: an equal-weight portfolio of N mid-tier creators, each with mean
# net return MU_M and standard deviation SD_M per yuan of budget, every
# pair correlated at RHO. Portfolio variance is
# SD^2 / N + (1 - 1/N) * RHO * SD^2,
# which falls towards RHO * SD^2 (the average covariance), never to zero.
# All parameters are illustrative (示意), not measured data.
import math
BUDGET = 600000 # 季度投放预算(元,示意)
MU_H, SD_H = 0.30, 1.20 # 顶流:每元预算的净回报均值与标准差(示意)
MU_M, SD_M = 0.24, 0.80 # 单名中腰部达人(示意)
RHOS = (0.0, 0.1, 0.2, 0.4)
def port_var(n, rho, sd=SD_M):
return sd * sd / n + (1.0 - 1.0 / n) * rho * sd * sd
print("budget %d yuan; head creator: mean %.2f, sd %.2f per yuan"
% (BUDGET, MU_H, SD_H))
print("sd of an equal-weight portfolio of N mid-tier creators")
print(" N" + "".join(" rho=%.2f" % r for r in RHOS))
for n in (1, 2, 5, 10, 20, 50, 100):
print("%5d" % n + "".join("%10.4f" % math.sqrt(port_var(n, r))
for r in RHOS))
print("limit" + "".join("%10.4f" % math.sqrt(r * SD_M * SD_M)
for r in RHOS))
print("N needed to come within 10% of the limit variance:"
+ "".join(" %d" % math.ceil((1 - r) / (0.1 * r)) for r in RHOS[1:]))
v20 = port_var(20, 0.2)
print("N=20, rho=0.20: variance %.4f, ratio to head variance %.4f"
% (v20, v20 / (SD_H * SD_H)))
print("sd in yuan: head %d, N=20 portfolio %d"
% (round(SD_H * BUDGET), round(math.sqrt(v20) * BUDGET)))
print("expected net return in yuan: head %d, N=20 portfolio %d"
% (round(MU_H * BUDGET), round(MU_M * BUDGET)))
print("diversifiable share of one creator's variance at rho=0.20: %.2f"
% (1 - 0.2))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《达人矩阵头部集中度与中腰部组合套利、尾部方差夏普比率与百名垂直达人投资组合母典》第 1 章「同一笔预算:顶流的方差与中腰部组合的方差公式」(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-276 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026276,
author = {Fan, Yafu},
title = {Head Concentration versus Mid-Tier Portfolios in Creator Matrices: Tail Variance, Sharpe Ratios and the Allocation of a Budget across a Hundred Niche Creators},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}