Forensic Accounting Toolkit (Benford, patterns, tracing, reconciliation)

What it is

A fraud-examiner toolkit over a transaction population: (a) a cited Benford first-digit screen on transaction amounts (observed vs P(d)=log10(1+1/d), chi-square vs the df=8 critical value 15.51, and MAD with Nigrini's 2012 bands); (b) pattern detection (round-number frequency, just-below-threshold clustering for structuring under the $10k BSA reporting rule, and duplicate amounts); (c) asset tracing through an entity chain (net position per entity, sinks vs sources, multi-hop flows); and (d) two-ledger reconciliation to the cent. Tracing, reconciliation and pattern counts are integer/exact (EXACT verdict); Benford is a statistic (numericTrust with a chi-square cross-check). HONEST: an anomaly is a flag for investigation, not proof; Benford is meaningless on assigned/bounded numbers and small samples; tracing sees only entered transfers. This complements the standalone Benford plugin - here Benford is one focused tool among several. Batch CSV in, per-tool table + chart out, CSV export. Deterministic, in-browser, no API/key/network/AI.

Honest scope

Authorities cited

• Benford, F. (1938). "The Law of Anomalous Numbers." Proc. Am. Philos. Soc. 78(4):551-572 - the first-digit law P(d)=log10(1+1/d). https://www.jstor.org/stable/984802
• Nigrini, M. J. (2012). Benford's Law: Applications for Forensic Accounting, Auditing, and Fraud Detection. Wiley - the Mean Absolute Deviation (MAD) conformity bands and small-sample guidance.
• Pearson chi-square goodness-of-fit: X^2 = sum (observed - expected)^2 / expected; df=8 critical value at alpha=0.05 is 15.507. Standard chi-square tables.
• 31 U.S.C. 5324 and 31 CFR 1010.311 - currency-transaction reporting at the $10,000 threshold; deliberately keeping amounts just below it is structuring. https://www.law.cornell.edu/uscode/text/31/5324