What does Bumps, Upsells and Cross-Sells conclude, in numbers, about “Cross-Fleet Routing by Conditional Mutual Information”?

Answer: Chapter 3 routes each of 30 synthetic sites to its top two by conditional mutual information and finds that even with 60,000 users only 14 to 17 sites have stable top twos.

Scope of the numbers: Valuation distributions, acceptance and refund rates and the needs structure of 30 sites are illustrative or synthetic; all data are generated by its programs; the book is not legal advice. Chapter 3 routes each of 30 synthetic sites to its top two by conditional mutual information and finds that even with 60,000 users only 14 to 17 sites have stable top twos.

中文版

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: 14, 17, 30, 60000

users sites with the true top-two set (of 30)
2,00010, 14, 10
10,00014, 12, 10
60,00016, 17, 14
bias, independent pair, n =2,000: 0.000361 bits
bias, independent pair, n = 10,000: 0.000072 bits
bias, independent pair, n = 60,000: 0.000012 bits
Show the full program
# Chapter 3. p06: How many users before the top-two routes are stable?
# Same synthetic model as p05. "Truth" = routes estimated from 400,000
# simulated users. For smaller samples (independent seeds) we count the
# share of the 30 sites whose estimated top-two set equals the true one,
# and the mean upward bias of the plug-in mutual information of an
# independent pair (the Miller-Madow bias, about 1/(2 n ln 2) bits).
import importlib.util
from math import log

def load(name):
    spec = importlib.util.spec_from_file_location(name, name + ".py")
    m = importlib.util.module_from_spec(spec)
    spec.loader.exec_module(m)
    return m

m = load("p05_cross_sell_mi")

if __name__ == "__main__":
    cnt, pair = m.estimate(400000, 1)
    truth = m.routes(cnt, pair, 400000)
    print("   users     sites with the true top-two set   (of 30)")
    for n in (2000, 10000, 60000):
        hits = []
        for seed in (11, 12, 13):
            c, p = m.estimate(n, seed)
            r = m.routes(c, p, n)
            hits.append(sum(set(r[i]) == set(truth[i]) for i in range(30)))
        print("   {:6,}    {}".format(n, ", ".join(str(h) for h in hits)))
    for n in (2000, 10000, 60000):
        print("   bias, independent pair, n = {:6,}: {:.6f} bits".format(
            n, 1 / (2 * n * log(2))))

Single source and citation

Fan, Y. (2026). Order-Bump Friction Elasticity, One-Click Upsell Kinetics and Cross-Domain Recommendation Matrices (Version 1.0.0), Chapter 3: Cross-Fleet Routing by Conditional Mutual Information. 5D Library. https://doi.org/10.5281/zenodo.24090207

DOI: 10.5281/zenodo.24090207 · Read the monograph · 5DL-B-078 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026078,
  author    = {Fan, Yafu},
  title     = {Order-Bump Friction Elasticity, One-Click Upsell Kinetics and Cross-Domain Recommendation Matrices},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  doi       = {10.5281/zenodo.24090207},
  url       = {https://doi.org/10.5281/zenodo.24090207}
}

Other chapter answers in this book