关于「东西方原典考辨:四柱结算法、纽科姆的对数表与帕乔利的试算平衡」,《本福特定律与流水报销审计》给出的定量结论是什么?

结论:第七章考辨四柱结算法、纽科姆 1881 年的短文与帕乔利 1494 年的复式记账。

数据口径:书中参数为示意;全部计算由书中程序生成;本书不是法律意见,也不是个人化经营建议,检验结果只是复核信号。 书中以标注框区分已发表的证据、可运行的 Python 计算(20 段程序)、原典与作者的假说。

English version

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

cells total identity chain count neither
196 96 0 0 0
2992 848 112 8 24
34608 4608 0 0 0
412160 10504 1500 24 132
silent alterations by family (prime = the following month):
gross-up, only X and K move68, e.g. +K +X
carried, S and J' move too88, e.g. +K -S -J' +X'
查看完整验算程序
# Chapter 7. p71: what the four-pillar identity can and cannot catch.
# A year of 12 monthly four-pillar statements.  Fields per month:
#   J old balance (jiu guan), X new receipts (xin shou),
#   K disbursements (kai chu), S actual balance (shi zai).
# Checks: identity J + X - K - S = 0 in every month; the chain
# J(m+1) = S(m) between consecutive months; and a cash count that
# anchors the opening J of month 1 and the closing S of month 12.
# An alteration adds +d or -d to between 1 and 4 cells inside two
# consecutive months; we enumerate every such alteration once and
# classify it by the first check that sees it.  Exact enumeration.
import itertools

FIELDS = "JXKS"
COEF = {"J": 1, "X": 1, "K": -1, "S": -1}
MONTHS = 12


def caught(vec):
    by_month = {}
    for (m, f), c in vec:
        by_month.setdefault(m, []).append(COEF[f] * c)
    for m, parts in by_month.items():
        tot = 0
        for p in parts:
            tot += p
        if tot != 0:
            return "identity"
    cells = dict(vec)
    for m in range(MONTHS - 1):
        if cells.get((m + 1, "J"), 0) != cells.get((m, "S"), 0):
            return "chain"
    if cells.get((0, "J"), 0) or cells.get((MONTHS - 1, "S"), 0):
        return "count"
    return "neither"


seen = set()
table = {}
silent = {}
for m in range(MONTHS - 1):
    cells = [(mm, f) for mm in (m, m + 1) for f in FIELDS]
    for size in (1, 2, 3, 4):
        for support in itertools.combinations(cells, size):
            for signs in itertools.product((1, -1), repeat=size):
                vec = tuple(sorted(zip(support, signs)))
                if vec in seen:
                    continue
                seen.add(vec)
                verdict = caught(vec)
                table[(size, verdict)] = table.get((size, verdict), 0) + 1
                if verdict == "neither":
                    moved = [f for (mm, f), c in vec]
                    carry = "J" in moved or "S" in moved
                    fam = "carry" if carry else "gross"
                    silent.setdefault(fam, []).append(vec)

VERDICTS = ("identity", "chain", "count", "neither")
print("%5s %8s %8s %8s %8s %8s" % (("cells", "total") + VERDICTS))
for size in (1, 2, 3, 4):
    row = [table.get((size, v), 0) for v in VERDICTS]
    tot = 0
    for r in row:
        tot += r
    print("%5d %8d %8d %8d %8d %8d" % tuple([size, tot] + row))


def show(vec):
    first = vec[0][0][0]
    return " ".join("%s%s%s" % ("+" if c > 0 else "-", f,
                                "'" if mm > first else "")
                    for (mm, f), c in vec)


print("silent alterations by family (prime = the following month):")
print("  gross-up, only X and K move: %d, e.g. %s"
      % (len(silent["gross"]), show(silent["gross"][0])))
print("  carried, S and J' move too:  %d, e.g. %s"
      % (len(silent["carry"]),
         [show(v) for v in silent["carry"]
          if show(v).startswith("+K -S")][0]))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《本福特首位数定律、账目舞弊检测与代购流水报销卡方审计母典》第 7 章「东西方原典考辨:四柱结算法、纽科姆的对数表与帕乔利的试算平衡」(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-216 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026216,
  author    = {Fan, Yafu},
  title     = {Benford's First-Digit Law, Ledger Fraud Detection and Chi-Square Audits of Buying-Agent Cash Flows and Expense Claims},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论