What does Pre-Owned Luxury Authentication and the Lemon Discount conclude, in numbers, about “Double Counting: How Naive Bayes Counts One Piece of Evidence Several Times When Features Share a Cause”?
Answer: Chapter 2 shows that when features share a cause, plain multiplication counts evidence twice: in a model whose marginal parameters are exactly right, one pattern has a naive posterior of 0.88503 and an exact posterior of only 0.34611, and a tempering weight fitted on confirmed records is 0.75.
Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not a pricing opinion on any actual transaction and not personalised legal or investment advice. Marked boxes separate published evidence, runnable Python calculations (20 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: 0.34611, 0.88503
| careful-tier rates 0.575 0.400 0.300 0.670 | |||||
| untouched-auth rates 0.0250 0.0375 0.0750 0.0225 | |||||
| phi(F2,F3): within counterfeits 0.267, within authentic 0.175 | |||||
| pattern | naive P | exact P | naive dB | exact dB | overcount |
| 0000 | 0.00057 | 0.00355 | -22.9 | -14.9 | -8.0 |
| 0110 | 0.09197 | 0.02915 | -0.4 | -5.7 | 5.3 |
| 1100 | 0.47523 | 0.30685 | 9.1 | 6.0 | 3.1 |
| 1110 | 0.88503 | 0.34611 | 18.4 | 6.8 | 11.6 |
| 1111 | 0.99929 | 0.99389 | 41.0 | 31.7 | 9.4 |
| 0111 | 0.94887 | 0.69907 | 22.2 | 13.2 | 9.0 |
Show the full program
# Chapter 2. p21: when features share a hidden cause, naive Bayes counts
# the same evidence twice. Counterfeits come from a crude workshop
# tier (weight 0.6) or a careful tier (0.4); authentic bags are either
# untouched (0.8) or refurbished (0.2), and refurbishing makes honest
# bags show anomalies. Within a tier features are independent, so the
# exact posterior is a two-component mixture. The tier rates are
# illustrative and chosen so that the marginal sensitivity and
# specificity equal those of p11 exactly.
import math
SENS = [0.80, 0.70, 0.60, 0.85]
FP = [0.05, 0.10, 0.15, 0.03] # 1 - specificity
CRUDE = [0.95, 0.90, 0.80, 0.97]
CAREFUL = [(s - 0.6 * c) / 0.4 for s, c in zip(SENS, CRUDE)]
REFURB = [0.15, 0.35, 0.45, 0.06]
PLAIN = [(m - 0.2 * r) / 0.8 for m, r in zip(FP, REFURB)]
FAKE = [(0.6, CRUDE), (0.4, CAREFUL)]
AUTH = [(0.2, REFURB), (0.8, PLAIN)]
PRIOR = 0.10
def lik(mix, pat):
return math.fsum(w * math.prod(r if b else 1 - r
for b, r in zip(pat, rates))
for w, rates in mix)
def corr(mix, i, j):
pi = lik_marg(mix, i)
pj = lik_marg(mix, j)
pij = math.fsum(w * rates[i] * rates[j] for w, rates in mix)
return (pij - pi * pj) / math.sqrt(pi * (1 - pi) * pj * (1 - pj))
def lik_marg(mix, i):
return math.fsum(w * rates[i] for w, rates in mix)
print("careful-tier rates", " ".join("{:.3f}".format(x) for x in CAREFUL))
print("untouched-auth rates", " ".join("{:.4f}".format(x) for x in PLAIN))
print("phi(F2,F3): within counterfeits {:.3f}, within authentic {:.3f}"
.format(corr(FAKE, 1, 2), corr(AUTH, 1, 2)))
print("pattern naive P exact P naive dB exact dB overcount")
po = PRIOR / (1 - PRIOR)
for s in ("0000", "0110", "1100", "1110", "1111", "0111"):
pat = [int(c) for c in s]
naive = math.prod((se / fp) if b else (1 - se) / (1 - fp)
for b, se, fp in zip(pat, SENS, FP))
exact = lik(FAKE, pat) / lik(AUTH, pat)
pn, pe = po * naive / (1 + po * naive), po * exact / (1 + po * exact)
dn, de = 10 * math.log10(naive), 10 * math.log10(exact)
print("{} {:.5f} {:.5f} {:6.1f} {:6.1f} {:5.1f}"
.format(s, pn, pe, dn, de, dn - de))Single source and citation
Fan, Y. (2026). Multi-Feature Bayesian Authentication of Pre-Owned Luxury, Akerlof Lemon-Discount Equilibria and the Resale Half-Life of Limited Bags (Version 1.0.0), Chapter 2: Double Counting: How Naive Bayes Counts One Piece of Evidence Several Times When Features Share a Cause. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-171 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026171,
author = {Fan, Yafu},
title = {Multi-Feature Bayesian Authentication of Pre-Owned Luxury, Akerlof Lemon-Discount Equilibria and the Resale Half-Life of Limited Bags},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Authentication as Bayesian Updating: Likelihood Ratios of Four Features, Channel Base Rates and Sequential Reading
- Chapter 3 · Agreement Between Appraisers: Cohen's Kappa, the Two Paradoxes of High Agreement and Low Kappa, and the Ceiling on Majority Votes
- Chapter 4 · The Lemon Discount: Why Owners of Genuine Bags Leave All at Once When the Share of Fakes Crosses a Tipping Point
- Chapter 5 · Third-Party Authentication and Refund Guarantees: How Low a Fee and How High a Refund Multiple Restore Equilibrium
- Chapter 6 · The Half-Life of a Limited Bag's Premium: Log-Linear Regression, Contamination by Fakes, and When a Bundle Quota Makes the Premium Collapse
- Chapter 7 · Reading the Sources: the Market Officer's Contracts and Claim Periods, the Man of Zheng Who Trusted His Measure, and Akerlof 1970
- Chapter 8 · Practice and Falsification: a Buy-In Quote Calculator, Machine Checks on Copy, and a Pre-Registered Test