5D Library › Two-Scale Renormalization, Occupancy Entropy and Thrust-Wave Superposition in Discrete Cyclical Networks · 中文
Chapter 5 Cross-Domain “Structural Isomorphism” Hypotheses: Sector Rotation, Flexible Vacancies, Twelve-Tone Equal Temperament and Periodic Staking
Note
This chapter is not investment advice. Its financial and crypto-asset examples serve only to show the distance between “formal resemblance” and “testable hypothesis”. The book recommends no asset, sector, coin, exchange or trading method, and does not consider the models of Chapters 2 and 3 usable for trading. Any real investment can lose all of its capital, and with leverage or derivatives losses can exceed it.
The author’s framework proposes that the structures of Chapters 1 to 3 apply “isomorphically” to four fields: sector rotation among financial assets, flexible vacancies in organisational structure, the acoustic structure of twelve-tone equal temperament, and periodic staking in decentralised networks. “Isomorphism” has a strict meaning in mathematics: a one-to-one map between two structures that preserves all their operations. Applied to a real field, it can at most be a hypothesis: if certain real quantities are matched to the cells, particles and antipodal pairing of the model, then certain consequences of the model should appear in data. This chapter writes out each matching and what is still missing before it can be tested.
A theorem is not evidence
First, a logical point that is easily overlooked. The entropy–peak bounds of Chapter 2 hold for any allocation of a fixed total to 12 categories, whether the categories are cells, sectors, departments or time slots. “Finding” in real data that entropy and peak satisfy these bounds is therefore no evidence that the model describes reality, just as finding in real data is no evidence that arithmetic describes reality.
Calculation
The first part of the program tests on 100,000 randomly generated allocations over 12 categories, of very different shapes; it holds in every one. The second part builds a null model with no “siphon” mechanism at all: the weights of 12 sectors are a softmax of 12 independent first-order autoregressive noise series, with no rule for money flowing between them. It then asks whether the allocation entropy in one period is correlated with the total change in weights in the next. The noise scales are illustrative.
Calculation
# Part 1: the Chapter 2 bounds hold for ANY allocation over 12 slots. Part 2: a SYNTHETIC
# null market with no siphon mechanism (softmax of 12 independent AR(1) noises): does low
# entropy still "predict" the size of the next move? (Illustrative parameters.)
import math, random
K = 12
def entropy(w):
return -sum(x * math.log2(x) for x in w if x > 0)
def softmax(x):
m = max(x); e = [math.exp(v - m) for v in x]; s = sum(e)
return [v / s for v in e]
def corr(a, b):
ma, mb = sum(a) / len(a), sum(b) / len(b)
cov = sum((x - ma) * (y - mb) for x, y in zip(a, b))
return cov / math.sqrt(sum((x - ma) ** 2 for x in a) * sum((y - mb) ** 2 for y in b))
if __name__ == "__main__":
rng = random.Random(2026)
ok = 0
for _ in range(100000):
w = [rng.expovariate(1.0) ** rng.choice((1, 3, 6)) for _ in range(K)]
s = sum(w); w = [x / s for x in w]
ok += 2 ** -entropy(w) <= max(w) + 1e-12 # peak >= 2^(-H)
print(f"part 1: max share >= 2^(-H) holds in {ok:,} of 100,000 random allocations")
print("part 2: noise corr(H_t, move_t+1) next move, lowest-H / highest-H fifth")
for sigma in (0.5, 1.0, 2.0):
x = [0.0] * K; hs, moves = [], []; w_old = softmax(x)
for t in range(20000):
x = [0.9 * v + sigma * math.sqrt(1 - 0.81) * rng.gauss(0, 1) for v in x]
w = softmax(x)
if t > 100:
hs.append(entropy(w_old)); moves.append(sum(map(abs, (a - b for a, b
in zip(w, w_old)))))
w_old = w
lo = sorted(range(len(hs)), key=hs.__getitem__)[: len(hs) // 5]
hi = sorted(range(len(hs)), key=hs.__getitem__)[-len(hs) // 5:]
m_lo = sum(moves[i] for i in lo) / len(lo)
m_hi = sum(moves[i] for i in hi) / len(hi)
print(f" {sigma:4.1f} {corr(hs, moves):+.3f}"
f" {m_lo:.3f} / {m_hi:.3f}")
# part 1: max share >= 2^(-H) holds in 100,000 of 100,000 random allocations
# part 2: noise corr(H_t, move_t+1) next move, lowest-H / highest-H fifth
# 0.5 -0.025 0.171 / 0.168
# 1.0 +0.087 0.306 / 0.326
# 2.0 +0.332 0.381 / 0.595The first part illustrates “a theorem is not evidence”. The second deserves more attention: in a null model with no mechanism, the correlation between this period’s entropy and next period’s change moves from −0.025 to +0.087 to +0.332 as the noise scale grows. At noise scale 2.0 the mean change after the lowest-entropy fifth of periods (0.381) is clearly smaller than after the highest-entropy fifth (0.595), because once the weight is concentrated in one sector the saturation of the softmax makes it hard to move. So if someone finds in real data that “low entropy is followed by larger moves”, or by smaller ones, neither can be taken directly as evidence for the model; one must first say what a suitable null model would give.
1. “Siphoning” in financial sector rotation
The matching. The 12 cells correspond to 12 industry sectors fixed in advance, particles to shares of capital, vacancies to sectors holding very little capital; “siphoning” corresponds to money flowing out of one sector into another; and antipodal coupling requires an advance specification of which two sectors form an “antipodal” pair.
Evidence
Known related facts. There is published evidence of momentum at the industry level: Moskowitz and Grinblatt found in 1999 that industries that performed well over a past period tend on average to perform well over the following period, and that much of the momentum in individual stocks can be explained by industry momentum (Moskowitz and Grinblatt 1999). But many of the predictors “discovered” in finance do not survive correction for multiple testing: Harvey, Liu and Zhu catalogued 316 reported cross-sectional return factors in 2016 and argued that, given the multiple testing, the hurdle for a new factor should be raised to a -statistic of about 3.0, so that many published factors are likely false discoveries (Harvey et al. 2016).
What it would need. (1) The sector division and the antipodal pairing must be fixed before the data are seen; otherwise 12 sectors can be paired off in ways, and one that “looks right” can always be found. (2) The conservation of “particles” does not hold in markets: market values change with prices, and inflows and outflows are not equal, so one must say whether flow data or market-value shares are used and how changes in the total are handled. (3) The prediction must concern the future, with a single outcome written down in advance and compared with a null model like the one above. (4) Trading costs, liquidity and short-selling constraints make any statistical regularity hard to exploit in practice. This book does not test this hypothesis; the test of Chapter 8 addresses a different and simpler question.
2. Flexible vacancies in organisations
The matching. Cells correspond to 12 posts or functions, particles to staff, vacancies to unfilled posts or deliberately reserved flexible positions; antipodal coupling corresponds to “when a post is vacant, its paired post covers it at half capacity”.
Evidence
Known research on organisational slack. Nohria and Gulati studied the relation between “slack resources” and innovation across the departments of two multinational companies in 1996, and found an inverted U: with too little or too much slack, innovation is lower (Nohria and Gulati 1996). This is compatible with the intuition that “some vacancy is good”, but it concerns slack resources, not this book’s occupation–coupling model.
What it would need. (1) The pairing between posts (who covers for whom) must be a rule that exists in the organisation beforehand and can be recorded. (2) Output must be measurable post by post. (3) The “coupling gain from vacancies” must be separated from the “specialisation gain from concentrating staff”, which has nothing to do with antipodal pairing. (4) Chapter 3 showed that the model’s total flux does not depend on neighbour relations between posts, so if what matters in an organisation is cooperation between “neighbouring” departments, the model does not apply.
3. Twelve-tone equal temperament: really is the mathematics of music
Of the author’s four applications, this one has the strictest mathematical correspondence: under octave equivalence the twelve pitch classes form , transposition is addition, inversion is negation, and the “transposition–inversion group” they generate is precisely the dihedral group (Crans et al. 2009). 5D Library volumes 3 and 31 discussed resonance and architectural acoustics; here the subject is treated only from the point of view of .
Calculation
Twelve-tone equal temperament divides the octave (frequency ratio 2) into 12 equal semitones, each with frequency ratio . The cent is defined as , so an equal-tempered semitone is exactly 100 cents. The program compares each equal-tempered interval with the just interval of simple whole-number ratio, computes the Pythagorean and syntonic commas, and spells out the algebraic meaning of the circle of fifths and of the antipode.
Calculation
# Twelve-tone equal temperament (12-TET) as arithmetic on Z12. Cents: 1200 * log2(ratio).
# Questions: how far is each 12-TET interval from a small-integer (just) ratio; why does
# the circle of fifths visit all twelve pitch classes; what is the "antipode" i + 6?
import math
from math import gcd
def cents(r):
return 1200 * math.log2(r)
JUST = [("minor second", 1, "16/15"), ("major second", 2, "9/8"),
("minor third", 3, "6/5"), ("major third", 4, "5/4"), ("fourth", 5, "4/3"),
("tritone", 6, "45/32"), ("tritone", 6, "7/5"), ("fifth", 7, "3/2"),
("minor sixth", 8, "8/5"), ("major sixth", 9, "5/3"),
("minor seventh", 10, "9/5"), ("major seventh", 11, "15/8")]
if __name__ == "__main__":
st = 2 ** (1 / 12)
print(f"semitone ratio 2^(1/12) = {st:.7f} = {cents(st):.0f} cents")
print("interval steps just just cents 12-TET 12-TET - just")
for name, k, frac in JUST:
p, q = map(int, frac.split("/"))
d = 100 * k - cents(p / q)
print(f"{name:15s} {k:3d} {frac:6s} {cents(p / q):8.2f}"
f" {100 * k:6d} {d:+7.2f}")
pc = 1.5 ** 12 / 2 ** 7
print(f"Pythagorean comma (3/2)^12 / 2^7 = {pc:.5f} = {cents(pc):.2f} cents;"
f" syntonic comma 81/80 = {cents(81 / 80):.2f}")
print(f"12 equal fifths are exactly 7 octaves; 12 just fifths overshoot by"
f" {cents(1.5 ** 12 / 2 ** 7) / 12:.3f} cents per fifth")
gens = [k for k in range(1, 12) if gcd(k, 12) == 1]
print("steps that generate all 12 pitch classes:", gens)
print("circle of fifths:", [7 * j % 12 for j in range(12)])
print("i + 6 is the tritone, 2^(6/12) = sqrt 2 =", f"{2 ** 0.5:.5f}")
print("nonzero classes equal to their own inversion (k = -k mod 12):",
[k for k in range(1, 12) if k == (-k) % 12])
print(f"harmonic series: partial 3 vs 2^(19/12) = {2 ** (19 / 12):.4f};"
f" partial 5 vs 2^(28/12) = {2 ** (28 / 12):.4f}")
# semitone ratio 2^(1/12) = 1.0594631 = 100 cents
# interval steps just just cents 12-TET 12-TET - just
# minor second 1 16/15 111.73 100 -11.73
# major second 2 9/8 203.91 200 -3.91
# minor third 3 6/5 315.64 300 -15.64
# major third 4 5/4 386.31 400 +13.69
# fourth 5 4/3 498.04 500 +1.96
# tritone 6 45/32 590.22 600 +9.78
# tritone 6 7/5 582.51 600 +17.49
# fifth 7 3/2 701.96 700 -1.96
# minor sixth 8 8/5 813.69 800 -13.69
# major sixth 9 5/3 884.36 900 +15.64
# minor seventh 10 9/5 1017.60 1000 -17.60
# major seventh 11 15/8 1088.27 1100 +11.73
# Pythagorean comma (3/2)^12 / 2^7 = 1.01364 = 23.46 cents; syntonic comma 81/80 = 21.51
# 12 equal fifths are exactly 7 octaves; 12 just fifths overshoot by 1.955 cents per fifth
# steps that generate all 12 pitch classes: [1, 5, 7, 11]
# circle of fifths: [0, 7, 2, 9, 4, 11, 6, 1, 8, 3, 10, 5]
# i + 6 is the tritone, 2^(6/12) = sqrt 2 = 1.41421
# nonzero classes equal to their own inversion (k = -k mod 12): [6]
# harmonic series: partial 3 vs 2^(19/12) = 2.9966; partial 5 vs 2^(28/12) = 5.0397The equal-tempered semitone ratio is 1.0594631. The just fifth 3/2 is 701.96 cents and the equal-tempered fifth 700 cents, 1.96 cents narrower; the just major third 5/4 is 386.31 cents, and the equal-tempered one is 13.69 cents wider; the just minor third 6/5 exceeds the equal-tempered one by 15.64 cents. Twelve just fifths overshoot seven octaves by 23.46 cents (the Pythagorean comma), and equal temperament spreads this excess evenly over the twelve fifths, about 1.955 cents each.
Algebraically, the step 7 (a fifth) is coprime to 12, so rising by fifths from any note passes through all 12 pitch classes: this is the circle of fifths. In the only steps that can do this are 1, 5, 7 and 11. The antipode of the ring is the tritone, with frequency ratio , the only non-zero interval equal to its own inversion; it differs from the just 45/32 by 9.78 cents and from 7/5 by 17.49 cents.
Evidence
The history of equal temperament and the physics of resonance. In the Ming dynasty, Zhu Zaiyu gave twelve-tone equal temperament by precise calculation in his Lüxue xinshuo of 1584 and set it out more systematically in the Lülü jingyi of 1595–96 (Kuttner 1975). Physically, the resonances of strings and pipes give the harmonic series, in which each partial’s frequency is a whole-number multiple of the fundamental and neighbouring partials stand in simple whole-number ratios. Twelve-tone equal temperament is not the set of natural frequencies of any resonant cavity, but a compromise that spreads the commas of just intonation evenly so that every key is available for modulation (Benson 2006). The last line of the program shows that the 3rd harmonic (3 times the fundamental) is close to, but not equal to, the equal-tempered note 19 semitones up (2.9966 times), and the 5th harmonic (5 times) differs more from the note 28 semitones up (5.0397 times).
The framework’s “acoustic resonant cavity of twelve-tone equal temperament” therefore needs correcting: the correspondence of equal temperament with and is a strict mathematical fact, but it is not a resonance phenomenon; nor does the antipodal coupling rule have any known counterpart in music (the tritone does not “absorb” half the energy of the pitch class opposite).
4. Periodic staking in decentralised networks
The matching. The author matches cells to 12 time slots in a cycle (for example 12 months), particles to locked (staked) shares, vacancies to slots without locked shares or just after unlocking, and antipodal coupling to some compensation rule between two slots half a cycle apart.
What it would need. (1) The matching holds strictly only when a network protocol has written in advance that rewards are allocated over 12 slots with antipodal compensation; it is then a design of the protocol, not a regularity “discovered”. (2) If the protocol has no such rule, an economic mechanism for antipodal coupling must be stated and tested on pre-registered on-chain data. (3) Staking carries the risks of smart-contract bugs, inability to exit during a lock-up, violent token price swings and regulatory change; all the capital can be lost. This book evaluates no specific protocol or token.
Hypothesis · not established
Author’s hypothesis: structural isomorphism (not yet established). The author conjectures that in at least one of the fields above there is a matching, writable in advance, under which the antipodal-pair model of Chapter 3 predicts a single outcome out of sample better than a null model and a neighbour-coupling model with the same number of parameters. A testable prediction must state five things before the data are seen: the division into 12 categories, the six antipodal pairs, the definition of particles and of the total, the single outcome, and the null model used for comparison. If, in a pre-registered test, the out-of-sample prediction error of the antipodal-pair model is no lower than that of the null model (the 95 % confidence interval for the difference includes 0 or is positive), the hypothesis is refuted in that field. Twelve-tone equal temperament is a mathematical fact and falls outside this hypothesis.
Chapter summary
The theorems of Chapters 2 and 3 hold for every allocation over 12 categories, so “finding” them in data is no evidence; a null model with no mechanism at all moves the correlation between entropy and next-period change from −0.025 to +0.332. Sector rotation and organisational vacancies can be tested only with categories, pairings, a definition of the conserved quantity, a single outcome and a null model all fixed in advance; the known findings on industry momentum and organisational slack are not this book’s model. The correspondence of twelve-tone equal temperament with and is a strict mathematical fact: the fifth is 1.96 cents narrow, the major third 13.69 cents wide, and the tritone is the antipode, ; but it is not the product of a resonant cavity. Periodic staking corresponds strictly only when the protocol itself is designed that way, and it carries the risk of losing all capital. The next chapter discusses why people so often see regularities where there are none.
References
Fan, Y. (2026). Two-Scale Renormalization, Occupancy Entropy and Thrust-Wave Superposition in Discrete Cyclical Networks: Algebraic Topology and Monte Carlo Verification on a Twelve-Node Ring Lattice [离散周期网络中的双尺度重整化、占位数分布熵与推力波叠加定理] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23796761 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).