What does Debiasing Collectible Returns conclude, in numbers, about “The Bailey-Muth-Nourse Regression: from Pairs of Sales to a Price Index”?
Answer: When a work's own volatility is only 0.10 the two conventions differ by 0.0053; after ten years the median work is worth 1.84 times its price and the mean work 1.94 times, and the chance of a loss on a single work is only 0.027. At an own volatility of 0.30 the arithmetic rate is 0.1119, which is 0.0489 above the geometric one; after ten years the mean work is worth 2.89 times its price, yet the median work is still worth only 1.84 times, and the chance that a single work is below its purchase price is 0.260. At a volatility of 0.50 the arithmetic rate is 0.2045 while the geometric rate stays at 0.0630, and the chance of a loss on one work is 0.350.
Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; it is not personalised legal or financial advice, and fee schedules and rules are subject to manual review before release. 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: 0.0053, 0.027, 0.0489, 0.0630, 0.10, 0.1119
| log drift m = 0.0611; horizon 10 years | ||||||
| s | median/y | mean/y | gap | med x10 | mean x10 | P(loss) |
| 0.10 | 0.0630 | 0.0683 | 0.0053 | 1.84 | 1.94 | 0.027 |
| 0.20 | 0.0630 | 0.0845 | 0.0215 | 1.84 | 2.25 | 0.167 |
| 0.30 | 0.0630 | 0.1119 | 0.0489 | 1.84 | 2.89 | 0.260 |
| 0.40 | 0.0630 | 0.1515 | 0.0885 | 1.84 | 4.10 | 0.315 |
| 0.50 | 0.0630 | 0.2045 | 0.1415 | 1.84 | 6.43 | 0.350 |
| half the variance, s 0.25 / 0.30 / 0.35: 0.0312 / 0.0450 / 0.0612 | ||||||
| s that turns a 0.063 median into a 0.087 mean: 0.211 |
Show the full program
# Chapter 2. p23: arithmetic index, geometric index and the typical piece
# Model: one piece's log price change over h years is normal with mean
# m*h and variance s*s*h. A log (geometric) repeat-sales index estimates
# m; the median piece grows at exp(m) - 1 a year, while the mean piece
# grows at exp(m + s*s/2) - 1. The drift m = log(1.063) takes the corrected
# Korteweg et al. figure as a log rate for illustration; the idiosyncratic
# volatilities s are illustrative. Closed form, no randomness.
import math
M = math.log(1.063)
H = 10
def phi(x):
return 0.5 * math.erfc(-x / math.sqrt(2.0))
def main():
print("log drift m = {:.4f}; horizon {} years".format(M, H))
print("{:>6}{:>10}{:>10}{:>8}{:>10}{:>10}{:>9}".format(
"s", "median/y", "mean/y", "gap", "med x10", "mean x10", "P(loss)"))
for s in (0.10, 0.20, 0.30, 0.40, 0.50):
med = math.expm1(M)
mean = math.expm1(M + 0.5 * s * s)
mx_med = math.exp(M * H)
mx_mean = math.exp(M * H + 0.5 * s * s * H)
ploss = phi(-M * H / (s * math.sqrt(H)))
print("{:>6.2f}{:>10.4f}{:>10.4f}{:>8.4f}{:>10.2f}{:>10.2f}"
"{:>9.3f}".format(s, med, mean, mean - med, mx_med, mx_mean,
ploss))
print("half the variance, s 0.25 / 0.30 / 0.35: {:.4f} / {:.4f} / "
"{:.4f}".format(0.5 * 0.25 ** 2, 0.5 * 0.30 ** 2, 0.5 * 0.35 ** 2))
s_needed = math.sqrt(2.0 * (math.log(1.087) - M))
print("s that turns a 0.063 median into a 0.087 mean: {:.3f}".format(
s_needed))
if __name__ == "__main__":
main()Single source and citation
Fan, Y. (2026). Repeat-Sales Selection Bias in Collectibles, Forgery Priors and the Illiquidity Discount of Alternative Assets (Version 1.0.0), Chapter 2: The Bailey-Muth-Nourse Regression: from Pairs of Sales to a Price Index. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-208 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026208,
author = {Fan, Yafu},
title = {Repeat-Sales Selection Bias in Collectibles, Forgery Priors and the Illiquidity Discount of Alternative Assets},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · After the Correction: Selection Bias in Repeat-Sales Samples
- Chapter 3 · Who Sells When: Resale Timing, Exit and Censoring
- Chapter 4 · Forgery Priors: Likelihood Ratios of Authentication Evidence and the Value of Information
- Chapter 5 · The Round Trip: Break-Even Years and the Chance of Loss on a Single Piece
- Chapter 6 · The Illiquidity Discount: a Perfect-Timing Bound and Sales under a Deadline
- Chapter 7 · Classics East and West: Mi Fu's History of Painting and Vasari's Lives
- Chapter 8 · Closing the Loop: an Allocation Cap, a Due-Diligence Checklist and a Falsification Plan