What does First-Move Probes and Hidden Market Truth conclude, in numbers, about “Identifiability: What the Back Door, the Instrument and the Published Threshold Each Require”?

Answer: Case B is the more common situation. When the proxy's accuracy a is 0.5 it carries no information about intent, and the stratified contrast is still +0.0700, with the full bias of +0.1500 intact. At a = 0.7 the adjusted value is +0.0537 and the bias +0.1337. At a = 0.9 the adjusted value is -0.0098 and the bias +0.0702, still large. Only when a equals 1.0, that is when the proxy is intent itself, does the bias vanish. The more the proxy resembles the real confounder, the less bias remains, but as long as it is not identical the bias does not disappear. The channels, devices and time slots in a typical log are, for the most part, shadows of intent.

Scope of the numbers: Parameters are illustrative, calculations come from the book's programs, and all data are synthetic. This is general business analysis, not personalised legal, tax or investment advice; it promises no income.

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: -0.0098, 0.0537, 0.0700, 0.0702, 0.1337, 0.1500

true effect -0.080
Aconfounder recorded: naive +0.0700adjusted -0.0800
Bhidden confounder, stratify on a proxy of accuracy a
a = 0.5: naive +0.0700adjusted +0.0700bias +0.1500
a = 0.6: naive +0.0700adjusted +0.0661bias +0.1461
a = 0.7: naive +0.0700adjusted +0.0537bias +0.1337
a = 0.8: naive +0.0700adjusted +0.0303bias +0.1103
a = 0.9: naive +0.0700adjusted -0.0098bias +0.0702
a = 1.0: naive +0.0700adjusted -0.0800bias +0.0000
Crandomised price, control for a post-price variable:
unadjusted -0.0800'adjusted' -0.1165
Show the full program
# Chapter 2. p21: back-door adjustment works when the confounder is seen.
# Exact enumeration over small discrete models (illustrative parameters).
# Purchase probability: 0.10 + 0.25 * (segment) - 0.08 * (high price).
# A: the confounder Z (channel) is recorded -> stratify on Z.
# B: the confounder U is hidden; Z is a proxy that matches U with
#    accuracy a -> stratifying on Z leaves part of the bias.
# C: the price is randomised, but we "control for" a variable C that the
#    price itself moves (a post-treatment variable) -> bias appears.
import math

EFFECT = -0.08


def buy(u, x):
    return 0.10 + 0.25 * u - 0.08 * x


def contrast(joint, key):
    # joint: dict (u, z, x) -> prob; adjust for the coordinate 'key'
    def p_buy(cond):
        num = math.fsum(p * buy(u, x) for (u, z, x), p in joint.items()
                        if cond(u, z, x))
        den = math.fsum(p for (u, z, x), p in joint.items() if cond(u, z, x))
        return num / den
    naive = p_buy(lambda u, z, x: x == 1) - p_buy(lambda u, z, x: x == 0)
    parts = []
    for s in (0, 1):
        ps = math.fsum(p for k, p in joint.items() if k[key] == s)
        d1 = p_buy(lambda u, z, x: x == 1 and (u, z, x)[key] == s)
        d0 = p_buy(lambda u, z, x: x == 0 and (u, z, x)[key] == s)
        parts.append(ps * (d1 - d0))
    return naive, math.fsum(parts)


def model_b(acc):
    joint = {}
    for u in (0, 1):
        for z in (0, 1):
            pz = acc if z == u else 1 - acc
            for x in (0, 1):
                ph = 0.8 if u == 1 else 0.2
                joint[(u, z, x)] = 0.5 * pz * (ph if x else 1 - ph)
    return joint


if __name__ == "__main__":
    print("true effect {:+.3f}".format(EFFECT))
    a = model_b(1.0)
    n, adj = contrast(a, 0)
    print("A  confounder recorded: naive {:+.4f}  adjusted {:+.4f}"
          .format(n, adj))
    print("B  hidden confounder, stratify on a proxy of accuracy a")
    for acc in (0.5, 0.6, 0.7, 0.8, 0.9, 1.0):
        n, adj = contrast(model_b(acc), 1)
        b = adj - EFFECT
        b = 0.0 if abs(b) < 1e-12 else b
        print("   a = {:.1f}: naive {:+.4f}  adjusted {:+.4f}  bias {:+.4f}"
              .format(acc, n, adj, b))
    # C: x randomised; 'z' slot now holds C, moved by x and by u
    joint = {}
    for u in (0, 1):
        for x in (0, 1):
            pc = 0.2 + 0.4 * x + 0.3 * u
            for c in (0, 1):
                joint[(u, c, x)] = 0.25 * (pc if c else 1 - pc)
    n, adj = contrast(joint, 1)
    print("C  randomised price, control for a post-price variable:")
    print("   unadjusted {:+.4f}  'adjusted' {:+.4f}".format(n, adj))

Single source and citation

Fan, Y. (2026). High-Agency Active Inference, Intentional Perturbation Probes and Hidden Market-Truth Exploration (Version 1.0.0), Chapter 2: Identifiability: What the Back Door, the Instrument and the Published Threshold Each Require. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-395 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026395,
  author    = {Fan, Yafu},
  title     = {High-Agency Active Inference, Intentional Perturbation Probes and Hidden Market-Truth Exploration},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book