What does Elasticities across Scale conclude, in numbers, about “Stratified Regression across Tiers: How One Straight Line Hides a Drifting Elasticity”?

Answer: Chapter 6 covers stratified regression across tiers of a matrix and the gap between a cross-sectional exponent of 1.150 among peers and an own-path exponent of 0.892.

Scope of the numbers: Marked boxes separate published evidence, runnable Python calculations (21 programs), classical sources and the author's hypotheses.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.892, 1.150

operators 40, quarters 8, growth 6 percent a quarter
cross-section exponent (peers)1.150
within-operator exponent (own)0.892 (true 0.85)
gap from brand-size link0.258 (built in 0.30)
doubling forecast: peers 2.220 times, own path 1.856 times
Show the full program
# Chapter 6. p63: the peers' exponent is not your own exponent.
# Model (illustrative): 40 operators. Operator j starts with n0 sites
# (log-uniform 1 to 50) and grows 8 quarters by 6 percent a quarter.
# Within an operator conversions follow Y = A_j n^0.85 exp(e), e ~ N(0,
# 0.1). Larger operators have better-run brands: ln A_j = 3 + 0.3 ln n0
# + N(0, 0.2). The cross-section of all operators in the last quarter
# gives one exponent; the within-operator (fixed effects) regression,
# which removes each operator's mean, gives another. Seed 20261121.
import math
import random

rng = random.Random(20261121)


def slope(xs, ys):
    n = len(xs)
    mx, my = math.fsum(xs) / n, math.fsum(ys) / n
    sxx = math.fsum((x - mx) ** 2 for x in xs)
    return math.fsum((x - mx) * (y - my) for x, y in zip(xs, ys)) / sxx


cross_x, cross_y, wx, wy = [], [], [], []
for _ in range(40):
    n0 = math.exp(rng.uniform(0, math.log(50)))
    la = 3 + 0.3 * math.log(n0) + rng.gauss(0, 0.2)
    xs, ys = [], []
    for q in range(8):
        n = n0 * 1.06 ** q
        xs.append(math.log(n))
        ys.append(la + 0.85 * math.log(n) + rng.gauss(0, 0.1))
    cross_x.append(xs[-1])
    cross_y.append(ys[-1])
    mx, my = math.fsum(xs) / 8, math.fsum(ys) / 8
    wx.extend(x - mx for x in xs)
    wy.extend(y - my for y in ys)
bc, bw = slope(cross_x, cross_y), slope(wx, wy)
print("operators 40, quarters 8, growth 6 percent a quarter")
print("cross-section exponent (peers)     %.3f" % bc)
print("within-operator exponent (own)     %.3f   (true 0.85)" % bw)
print("gap from brand-size link           %.3f   (built in 0.30)"
      % (bc - bw))
print("doubling forecast: peers %.3f times, own path %.3f times"
      % (2 ** bc, 2 ** bw))

Single source and citation

Fan, Y. (2026). Renormalisation Group and Cross-Scale Scaling Laws: Which Conversion Elasticities Survive from a One-Person Workshop to a Hundred-Site Matrix (Version 1.0.0), Chapter 6: Stratified Regression across Tiers: How One Straight Line Hides a Drifting Elasticity. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-248 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026248,
  author    = {Fan, Yafu},
  title     = {Renormalisation Group and Cross-Scale Scaling Laws: Which Conversion Elasticities Survive from a One-Person Workshop to a Hundred-Site Matrix},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book