5D Library › Affiliate Network Power Laws, Influencer Portfolio Optimization and Incentive Stability Bounds · 中文

Chapter 1. Heavy-Tailed Output in Affiliate Networks and Price’s Law

When does “the square root of N produce half” hold?

The scientometrician Derek de Solla Price proposed that among NN authors, the most productive N\sqrt{N} write about half of the papers (Price 1963). This rule of thumb is often carried over to sales teams and affiliate networks: of ten thousand affiliates, the best hundred sell half. Whether it holds depends on how heavy the tail of output is. Below, each affiliate’s output follows a Lomax (Pareto type II) law with tail index α\alpha; the smaller α\alpha, the heavier the tail.

Calculation

# Chapter 1. p01: Does the top square root of N really produce half? Price's
# law says that in a group of N producers, the top sqrt(N) produce about
# half of all output. Here output per affiliate follows a Lomax (Pareto
# type II) law with tail index alpha: x = U^(-1/alpha) - 1. For several
# alpha and N, the median share of the top sqrt(N) over repeats is shown,
# seed 87. A smaller alpha means a heavier tail (alpha <= 1: infinite mean).
import random
from math import isqrt

def lomax(rng, alpha):
    return (1.0 - rng.random()) ** (-1.0 / alpha) - 1.0

def top_share(rng, n, alpha):
    xs = sorted((lomax(rng, alpha) for _ in range(n)), reverse=True)
    k = isqrt(n)
    return sum(xs[:k]) / sum(xs)

if __name__ == "__main__":
    rng = random.Random(87)
    sizes = (100, 1000, 10000)
    print("   share of output from the top sqrt(N) (median of repeats)")
    print("   alpha    N = 100   N = 1,000   N = 10,000")
    for alpha in (1.1, 1.5, 2.0, 3.0):
        row = []
        for n in sizes:
            reps = 300 if n < 10000 else 60
            s = sorted(top_share(rng, n, alpha) for _ in range(reps))
            row.append(s[reps // 2])
        print("   {:5.1f}    {:7.3f}   {:9.3f}   {:10.3f}"
              .format(alpha, *row))
    print("   top sqrt(N) as a share of people: {:.3f} / {:.3f} / {:.3f}"
          .format(10 / 100, 31 / 1000, 100 / 10000))
#    share of output from the top sqrt(N) (median of repeats)
#    alpha    N = 100   N = 1,000   N = 10,000
#      1.1      0.691       0.603        0.545
#      1.5      0.584       0.422        0.299
#      2.0      0.492       0.306        0.189
#      3.0      0.430       0.228        0.119
#    top sqrt(N) as a share of people: 0.100 / 0.031 / 0.010

The share of the top N\sqrt{N} comes near one half in only two combinations: NN = 100 with α\alpha = 2.0 (0.492), or the very heavy tail α\alpha = 1.1 (0.545 at NN = 10,000). At α\alpha = 1.5 the share falls from 0.584 for a hundred people to 0.299 for ten thousand. The reason is simple: N\sqrt{N} is a shrinking fraction of everyone as NN grows, only 0.010 at ten thousand; unless the tail is so heavy that a few can outweigh the whole, this small group cannot keep carrying half.

Output share of the top \sqrt{N} as the network grows (p01).

Dormant affiliates and estimating the tail index

In a real affiliate network, most registrants never make a sale. p02 sets 75% of ten thousand registrants as dormant, lets the active quarter sell according to a Lomax law with α\alpha = 1.5, and then estimates the tail index from the kk largest sellers with Hill’s estimator (Hill 1975; Clauset et al. 2009).

Calculation

# Chapter 1. p02: Dormant affiliates and the tail index. Of 10,000
# registered affiliates, 75% never make a sale; the active quarter sell
# 10 x Lomax(alpha = 1.5) units a year (see p01). The top sqrt(N) share
# depends on whether N counts everyone or only the active. The tail
# index is then estimated with Hill's estimator from the k largest
# sellers, alpha_hat = k / sum(log(x_i / x_(k+1))), for several k: too
# large a k reaches into the body of the law, where Lomax is not a pure
# power law, and biases alpha_hat downward: the tail looks heavier than
# it is. 200 repeats, seed 87.
import random
from math import isqrt, log

def network(rng, n=10000, dormant=0.75, alpha=1.5):
    out = []
    for _ in range(n):
        if rng.random() < dormant:
            out.append(0.0)
        else:
            out.append(10.0 * ((1.0 - rng.random()) ** (-1.0 / alpha) - 1))
    return out

def hill(xs_desc, k):
    ref = xs_desc[k]
    return k / sum(log(x / ref) for x in xs_desc[:k])

if __name__ == "__main__":
    rng = random.Random(87)
    ks = (25, 50, 100, 250, 500, 1000)
    est = {k: [] for k in ks}
    s_all, s_act, med, zero = [], [], [], []
    for _ in range(200):
        xs = sorted(network(rng), reverse=True)
        act = [x for x in xs if x > 0]
        s_all.append(sum(xs[:isqrt(len(xs))]) / sum(xs))
        s_act.append(sum(act[:isqrt(len(act))]) / sum(act))
        med.append(xs[len(xs) // 2])
        zero.append(1 - len(act) / len(xs))
        for k in ks:
            est[k].append(hill(xs, k))
    mid = lambda v: sorted(v)[len(v) // 2]
    print("   top sqrt(N) share, N = all registered (top 100): {:.3f}"
          .format(mid(s_all)))
    print("   top sqrt(N) share, N = active only (top {}):      {:.3f}"
          .format(isqrt(2500), mid(s_act)))
    print("   median registered affiliate sells {:.1f}; share never"
          " selling {:.3f}".format(mid(med), mid(zero)))
    print("   Hill estimate of alpha (true 1.5)")
    print("       k   median   5% to 95%")
    for k in ks:
        v = sorted(est[k])
        print("   {:5,d}   {:6.2f}   {:.2f} to {:.2f}".format(
            k, v[100], v[10], v[189]))
#    top sqrt(N) share, N = all registered (top 100): 0.460
#    top sqrt(N) share, N = active only (top 50):      0.356
#    median registered affiliate sells 0.0; share never selling 0.750
#    Hill estimate of alpha (true 1.5)
#        k   median   5% to 95%
#       25     1.50   1.10 to 2.22
#       50     1.45   1.16 to 1.87
#      100     1.39   1.17 to 1.63
#      250     1.30   1.17 to 1.44
#      500     1.17   1.09 to 1.25
#    1,000     0.97   0.92 to 1.01

Counting dormant registrants in NN, the top hundred hold a share of 0.460, which looks very like Price’s law; counting only the active, the top fifty hold 0.356. Same network, and the conclusion depends on the denominator. The median registrant’s sales are zero. The tail estimate is just as sensitive to kk: at kk = 25 the median estimate is 1.50 but the range is as wide as 1.10 to 2.22; with kk at 500 the estimate falls to 1.17, and at 1,000 to only 0.97, making the tail look heavier than it is and even crossing the line where the mean becomes infinite.

Note

Three rules when reporting concentration: state whether NN counts registrants or active affiliates; report the median as well as the mean; when estimating the tail index, give how kk was chosen and the range, not a single number.

Summary

Price’s law is not a constant but a function of the tail index and the size of the network; under a common heavy tail such as α\alpha = 1.5, the larger the network, the smaller the share of the top N\sqrt{N}. Dormant registrants make concentration look higher, and Hill’s estimator makes the tail look too heavy when kk is taken too large.

References

Clauset, Aaron, Cosma Rohilla Shalizi, and M. E. J. Newman. 2009. “Power-Law Distributions in Empirical Data.” SIAM Review 51 (4): 661 to 703.
Hill, Bruce M. 1975. “A Simple General Approach to Inference about the Tail of a Distribution.” The Annals of Statistics 3 (5): 1163 to 1174.
Price, Derek J. de Solla. 1963. Little Science, Big Science. Columbia University Press.

Fan, Y. (2026). Affiliate Network Power Laws, Influencer Portfolio Optimization and Incentive Stability Bounds: Heavy-Tailed Productivity, Markowitz Traffic Allocation and Single-Tier Commission Equilibria [分销网络幂律激活、网红矩阵组合优化与激励稳定边界] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090229 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).