What does Da Liu Ren as Group Action and Decision Tree conclude, in numbers, about “Practice and Falsification: a Style Library, a Pre-registered Test and the Conditions for Refutation”?
Answer: Chapter 8 gives an operator's checklist, design tokens for a front-end style library and a pre-registered archive test with a power of 0.895 at 200 cases. Marked boxes separate published evidence, runnable Python calculations (19 programs), classical sources and the author's hypotheses.
Scope of the numbers: Its parameters about users, archives and prices are illustrative; all calculations are generated by its programs; it analyses the structure and probabilities of a system, makes no personal reading, and is not legal, financial or business advice. Marked boxes separate published evidence, runnable Python calculations (19 programs), classical sources and the author's hypotheses.
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: 0.895, 200
| cells: 7 | effect size w = 0.307 | |
| biyong | H0 0.1125 | H1 0.0935 |
| fanyin | H0 0.0833 | H1 0.0693 |
| fuyin | H0 0.0833 | H1 0.0693 |
| rare | H0 0.0472 | H1 0.0785 |
| shehai | H0 0.1056 | H1 0.1755 |
| yaoke | H0 0.1000 | H1 0.1247 |
| zeike | H0 0.4681 | H1 0.3891 |
| n | critical chi2 | power |
| 50 | 12.81 | 0.348 |
| 100 | 12.66 | 0.603 |
| 150 | 12.59 | 0.790 |
| 200 | 12.92 | 0.895 |
| 300 | 12.77 | 0.981 |
| 400 | 12.57 | 0.998 |
| smallest n in the grid with power >= 0.80: 200 |
Show the full program
# Chapter 8. p81: Sample size for the pre-registered archive test. H0: the
# gates of recorded cases follow the exact shares of p31 (rare gates
# maoxing, bieze, bazhuan pooled, 7 cells). H1 (illustrative): a recorder
# keeps she-hai and rare cases twice as often and yao-ke 1.5 times as
# often as plain ones. Critical value and power are both simulated
# (seeded); alpha = 0.05, 4,000 draws per cell of the table. Cohen's
# conventions for the effect size w: 0.1 small, 0.3 medium, 0.5 large.
import math
import random
from collections import Counter
B = ("zi", "chou", "yin", "mao", "chen", "si", "wu", "wei", "shen",
"you", "xu", "hai")
G = ("jia", "yi", "bing", "ding", "wu", "ji", "geng", "xin", "ren", "gui")
BE = (4, 2, 0, 0, 2, 1, 1, 2, 3, 3, 2, 4) # 0 wood 1 fire 2 earth
GE = (0, 0, 1, 1, 2, 2, 3, 3, 4, 4) # 3 metal 4 water
GONG = (2, 4, 5, 7, 5, 7, 8, 10, 11, 1) # lodging palace of each stem
XING = (3, 10, 5, 0, 4, 8, 6, 1, 2, 9, 7, 11) # punishment map
def ke(a, b):
"""True if element a overcomes element b."""
return (b - a) % 5 == 2
def uniq(xs):
out = []
for x in xs:
if x not in out:
out.append(x)
return out
def depth(u, s, kind, cnt="A", home=True):
"""She-hai depth of upper u walking home from ground (u - s)."""
n = 0
for i in range(s + 1 if home else s):
q = (u - s + i) % 12
for e in [BE[q]] + [GE[g] for g in range(10) if GONG[g] == q]:
if cnt == "B" and kind == "ke":
n += ke(BE[u], e)
else:
n += ke(e, BE[u])
return n
def pick(pool, k, s, kind, v):
g = k % 10
if len(pool) == 1:
return pool[0], "one"
bi = [u for u in pool if u % 2 == g % 2]
if len(bi) == 1:
return bi[0], "biyong"
pool = bi or pool
tied = pool
if v.get("mode", "depth") == "depth":
ds = [depth(u, s, kind, v.get("cnt", "A"), v.get("home", True))
for u in pool]
tied = [u for u, d in zip(pool, ds) if d == max(ds)]
if len(tied) == 1:
return tied[0], "depth"
for lab, r in (("jianji", 2), ("chawei", 0)):
on = [u for u in tied if (u - s) % 12 % 3 == r]
if len(on) == 1:
return on[0], lab
if on:
break
z = GONG[g] if g % 2 == 0 else k % 12
return (z + s) % 12, "zhuixia"
def lessons(k, s):
a = (GONG[k % 10] + s) % 12
c = (k % 12 + s) % 12
return a, (a + s) % 12, c, (c + s) % 12
def sanchuan(k, s, v=None):
"""(initial, middle, final, gate, how) for day k and shift s."""
v = v or {}
g, z = k % 10, k % 12
yang = g % 2 == 0
a, b, c, d = lessons(k, s)
low = ((a, GE[g]), (b, BE[a]), (c, BE[z]), (d, BE[c]))
zei = uniq([u for u, e in low if ke(e, BE[u])])
kk = uniq([u for u, e in low if ke(BE[u], e)])
if s == 0:
gan = bool(zei or kk) or yang
x = a if gan else c
y = (c if gan else a) if XING[x] == x else XING[x]
w = (y + 6) % 12 if XING[y] == y else XING[y]
how = "withke" if zei or kk else ("yangday" if yang else "yinday")
return x, y, w, "fuyin", how
if zei or kk:
x, how = pick(zei or kk, k, s, "zei" if zei else "ke", v)
if how == "one":
how = "chongshen" if zei else "yuanshou"
gate = {"chongshen": "zeike", "yuanshou": "zeike",
"biyong": "biyong"}.get(how, "shehai")
if s == 6:
gate = "fanyin"
return x, (x + s) % 12, (x + 2 * s) % 12, gate, how
if s == 6:
sheng = [t % 12 for t in (z, z + 4, z + 8) if t % 3 == 2][0]
return (sheng + 6) % 12, c, a, "fanyin", "jinglan"
ups = uniq([b, c, d])
p1 = [u for u in ups if ke(BE[u], GE[g])]
p2 = [u for u in ups if ke(GE[g], BE[u])]
same = GONG[g] == z
if (p1 or p2) and not (same and v.get("bz_first")):
x, _ = pick(p1 or p2, k, s, "ke", v)
how = "haoshi" if p1 else "danshe"
return x, (x + s) % 12, (x + 2 * s) % 12, "yaoke", how
if same:
x = (a + 2) % 12 if yang else (d - 2) % 12
return x, a, a, "bazhuan", "bazhuan"
if len({(a, GONG[g]), (b, a), (c, z), (d, c)}) < 4:
x = (GONG[(g + 5) % 10] + s) % 12 if yang else (z + 4) % 12
return x, a, a, "bieze", "bieze"
if yang:
return (9 + s) % 12, c, a, "maoxing", "mao-yang"
return (9 - s) % 12, a, c, "maoxing", "mao-yin"
rng = random.Random(20261112)
POOL = {"maoxing": "rare", "bieze": "rare", "bazhuan": "rare"}
c = Counter(POOL.get(sanchuan(k, s)[3], sanchuan(k, s)[3])
for k in range(60) for s in range(12))
cells = sorted(c)
p0 = [c[x] / 720 for x in cells]
lift = {"shehai": 2.0, "rare": 2.0, "yaoke": 1.5}
raw = [p * lift.get(x, 1.0) for x, p in zip(cells, p0)]
p1 = [r / math.fsum(raw) for r in raw]
w = math.sqrt(math.fsum((b - a) ** 2 / a for a, b in zip(p0, p1)))
print("cells:", len(cells), " effect size w = {:.3f}".format(w))
for x, a, b in zip(cells, p0, p1):
print(" {:<8} H0 {:.4f} H1 {:.4f}".format(x, a, b))
def stat(n, probs):
got = Counter(rng.choices(range(len(cells)), weights=probs, k=n))
return math.fsum((got[i] - n * p0[i]) ** 2 / (n * p0[i])
for i in range(len(cells)))
print(" n critical chi2 power")
need = None
for n in (50, 100, 150, 200, 300, 400):
null = sorted(stat(n, p0) for _ in range(4000))
crit = null[int(0.95 * 4000)]
pw = len([1 for _ in range(4000) if stat(n, p1) > crit]) / 4000
if need is None and pw >= 0.8:
need = n
print("{:4d} {:13.2f} {:.3f}".format(n, crit, pw))
print("smallest n in the grid with power >= 0.80:", need)Single source and citation
Fan, Y. (2026). Da Liu Ren Heaven-Earth Plate Permutations, Four-Lesson Three-Transmission Digraphs and the Depth Algebra of the Nine Methods (Version 1.0.0), Chapter 8: Practice and Falsification: a Style Library, a Pre-registered Test and the Conditions for Refutation. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-153 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026153,
author = {Fan, Yafu},
title = {Da Liu Ren Heaven-Earth Plate Permutations, Four-Lesson Three-Transmission Digraphs and the Depth Algebra of the Nine Methods},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · The Plates: a Cyclic Permutation Group on Twelve Palaces
- Chapter 2 · The Four Lessons: Lodging Palaces, the Sixty-Day Cycle and Lesson Counts
- Chapter 3 · The Nine Methods as a Decision Tree: Exact Enumeration of 8,640 Situations
- Chapter 4 · The Depth Algebra of She-hai: Arc Sums, Saturation and Disagreement Between Schools
- Chapter 5 · The Three Transmissions as a Digraph: Rotation Orbits and Path Shapes
- Chapter 6 · Branch Probabilities and Information: Calendar Sampling, Mutual Information and Uniform Weights
- Chapter 7 · The Classics Examined: the Rengui, the Case Records of Shao Yanhe and Ptolemy's Tetrabiblos