What does Three Tiers, Pilot Deposits and 48-Hour Deals conclude, in numbers, about “Anchoring, Compromise and Decoy Effects Measured: When the Middle Tier Is Chosen”?

Answer: Chapter 3 measures anchoring, compromise and decoy effects, and under loss aversion adding the third tier raises the middle tier's share from 0.518 to 0.576;

Scope of the numbers: All parameters are illustrative and every calculation comes from 22 standard-library programs; deadlines must be real and never reset, marketing messages need prior consent and a one-click unsubscribe, and no multi-level commission is involved.

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: 0.518, 0.576

loss weight lam = 1.0
  menu A+B     A 0.480  B 0.520   USD 661.80 per chooser
  menu B+C     B 0.464  C 0.536   USD 1,727.17 per chooser
  menu A+B+C   A 0.334  B 0.276  C 0.390   USD 1,315.01 per chooser
  share of B when C is added: 0.520 -> 0.276 (falls)
loss weight lam = 2.0
  menu A+B     A 0.482  B 0.518   USD 660.69 per chooser
  menu B+C     B 0.460  C 0.540   USD 1,733.27 per chooser
  menu A+B+C   A 0.209  B 0.576  C 0.215   USD 1,136.12 per chooser
  share of B when C is added: 0.518 -> 0.576 (rises)
Show the full program
# Chapter 3. p31: the compromise effect in a loss-averse, reference-dependent choice model.
# Each buyer judges two attributes, benefit ln(1 + scope units) and price
# (as -ln price), against a reference point at the midpoint of the menu on
# each attribute; losses weigh lam times gains (extremeness aversion).
# Buyers differ in the weight they give benefit (lognormal, sd 0.5); the
# choice among options is logit with scale MU.  Seeded; illustrative only.
import math
import random

OPT = {"A": (4, 350.0), "B": (12, 950.0), "C": (32, 2400.0)}
MU = 1.6
N = 20000


def shares(menu, lam, seed=11):
    rng = random.Random(seed)
    att = {k: (math.log(1 + OPT[k][0]), -math.log(OPT[k][1])) for k in menu}
    ref = [0.5 * (min(a[j] for a in att.values()) + max(a[j] for a in att.values()))
           for j in (0, 1)]
    tot = {k: [] for k in menu}
    for _ in range(N):
        wb = math.exp(rng.gauss(0.0, 0.5))
        u = {}
        for k, (b, p) in att.items():
            parts = []
            for w, x, r in ((wb, b, ref[0]), (1.0, p, ref[1])):
                g = x - r
                parts.append(w * (g if g >= 0 else lam * g))
            u[k] = MU * math.fsum(parts)
        top = max(u.values())
        z = math.fsum(math.exp(v - top) for v in u.values())
        for k in menu:
            tot[k].append(math.exp(u[k] - top) / z)
    return {k: math.fsum(v) / N for k, v in tot.items()}


for lam in (1.0, 2.0):
    print("loss weight lam = {:.1f}".format(lam))
    for menu in (("A", "B"), ("B", "C"), ("A", "B", "C")):
        s = shares(menu, lam)
        rev = math.fsum(s[k] * OPT[k][1] for k in menu)
        print("  menu {:<7} ".format("+".join(menu)) + "  ".join(
            "{} {:.3f}".format(k, s[k]) for k in menu) + "   USD {:,.2f} per chooser".format(rev))
    two, three = shares(("A", "B"), lam)["B"], shares(("A", "B", "C"), lam)["B"]
    print("  share of B when C is added: {:.3f} -> {:.3f} ({})".format(
        two, three, "rises" if three > two else "falls"))

Single source and citation

Fan, Y. (2026). Productised B2B Service Packaging, Low-Friction Pilot Deposit Anchoring and Forty-Eight-Hour Deal Velocity (Version 1.0.0), Chapter 3: Anchoring, Compromise and Decoy Effects Measured: When the Middle Tier Is Chosen. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-352 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026352,
  author    = {Fan, Yafu},
  title     = {Productised B2B Service Packaging, Low-Friction Pilot Deposit Anchoring and Forty-Eight-Hour Deal Velocity},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book