Ground Walk Score conjecture (Duncan et al. 2011), freeze version date

- Added verified reference: Duncan, Aldstadt, Whalen, Melly & Gortmaker
  (2011), IJERPH 8(11), 4160-4179, DOI 10.3390/ijerph8114160 (verified
  via PubMed/MDPI) — validation of Walk Score across four US metros,
  supporting the market-specific-meaning mechanism in the discussion.
- Title page date frozen to August 5, 2026 (was \today).
- Build: 60 pages, 0 errors, 0 undefined refs, 0 overfull boxes,
  70/70 references cited.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

simon-pierre boucher committed 5 days ago (Aug 5, 2026) parent 1279445

Showing 6 changed files with +27 and −11

modified README.md +1 −1

		`@@ -18,7 +18,7 @@`
18	18	[![Sample](https://img.shields.io/badge/Sample-788%2C842%20listings-2B5C8A)](#-data)
19	19	[![Coverage](https://img.shields.io/badge/Coverage-50%20states%20%2B%20DC-2B5C8A)](#-data)
20	20	[![Figures](https://img.shields.io/badge/Figures-11%20publication--ready-A63E38)](#-figures)
21		−[![References](https://img.shields.io/badge/References-69%20verified%20(OpenAlex%2FDOI)-2E8B57)](paper/references.bib)
	21	+[![References](https://img.shields.io/badge/References-70%20verified%20(OpenAlex%2FDOI)-2E8B57)](paper/references.bib)
22	22	[![Verified](https://img.shields.io/badge/Results-verified%20against%20stored%20artifacts-2E8B57)](#-verification--integrity)
23	23	[![Status](https://img.shields.io/badge/Status-Working%20Paper%20v1.1-orange)](#-citation)
24	24

modified UPGRADE_REPORT.md +8 −6

		@@ -127,10 +127,12 @@ prediction; `bishop2020best`/`rosen1974hedonic` for the SHAP-is-not-WTP claim).
127	127	2. Campbell, Giglio & Pathak (2011) magnitude — ✅ verified against the AER
128	128	abstract: average foreclosure discount of 27\% on Massachusetts transactions.
129	129	Text unchanged (accurate).
130		−3. “Walk Score's metro-relative construction” conjecture — the claim that
131		− neighborhood scores carry market-specific scale is now explicitly flagged in
132		− the discussion as “plausible rather than established.” If you have a source on
133		− Walk Score's construction, it could be cited there.
	130	+3. “Neighborhood scores carry market-specific meaning” conjecture — ✅ now
	131	+ partially grounded: Duncan et al. (2011, IJERPH 8(11), 4160–4179, DOI
	132	+ 10.3390/ijerph8114160; verified via PubMed/MDPI) validate Walk Score across
	133	+ four US metropolitan areas and find its correspondence with walkability
	134	+ constructs varies by metro. Cited in the discussion; the mechanism remains
	135	+ flagged as “plausible rather than established” since we do not test it.
134	136	4. Positioning sentence — “To our knowledge, this framing has not previously
135	137	been brought to bear on national-scale hedonic housing models” (end of
136	138	literature §validation). Standard novelty hedge, but worth your sign-off.
		@@ -157,8 +159,8 @@ prediction; `bishop2020best`/`rosen1974hedonic` for the SHAP-is-not-WTP claim).
157	159	## 7. Build status & typographic fixes
158	160
159	161	`latexmk` clean build: **60 pages, 0 errors, 0 undefined references/citations,
160		−0 overfull vboxes**, 69/69 entries cited, hyperlinks resolving. Version 1.1,
161		−dated \today at compile time (freeze before submission if desired).
	162	+0 overfull vboxes**, 70/70 entries cited, hyperlinks resolving. Version 1.1; title-page
	163	+date frozen to August 5, 2026 for submission.
162	164
163	165	Two pre-existing typographic defects were found and fixed during the final QA
164	166	pass: the main OLS table (Table 5) and the appendix full-OLS landscape table

modified paper/main.pdf +0 −0

Binary file not shown.

modified paper/references.bib +11 −0

		`@@ -385,6 +385,17 @@ at-sign cannot appear here.`
385	385	doi = {10.1093/rfs/hhz134},
386	386	}
387	387
	388	+@article{duncan2011validation,
	389	+ author = {Duncan, Dustin T. and Aldstadt, Jared and Whalen, John and Melly, Steven J. and Gortmaker, Steven L.},
	390	+ title = {Validation of {W}alk {S}core for estimating neighborhood walkability: {A}n analysis of four {US} metropolitan areas},
	391	+ journal = {International Journal of Environmental Research and Public Health},
	392	+ year = {2011},
	393	+ volume = {8},
	394	+ number = {11},
	395	+ pages = {4160--4179},
	396	+ doi = {10.3390/ijerph8114160},
	397	+}
	398	+
388	399	@article{pivo2011walkability,
389	400	author = {Pivo, Gary and Fisher, Jeffrey D.},
390	401	title = {The walkability premium in commercial real estate investments},

modified paper/sections/discussion.tex +6 −3

		`@@ -122,9 +122,12 @@ held-out data.`
122	122
123	123	The ablation analysis identifies \emph{what} fails to transfer. Neighborhood
124	124	features contribute the largest random-validation gain ($+17.6$ pp) but only
125		−$+3.5$ pp under geographic holdout---consistent with our conjecture that scores
126		−like Walk Score and school ratings carry market-specific scaling; we flag this
127		−mechanism as plausible rather than established, since we do not test it directly.
	125	+$+3.5$ pp under geographic holdout---consistent with the conjecture that scores
	126	+like Walk Score and school ratings carry market-specific meaning. Validation
	127	+research on Walk Score itself finds that its correspondence with underlying
	128	+walkability constructs varies across metropolitan areas
	129	+\citep{duncan2011validation}, which supports the mechanism, though we do not
	130	+test it directly and flag it as plausible rather than established.
128	131	More striking, the features most obviously "about" geography are actively harmful
129	132	out-of-market: the full model with region dummies loses 7.6 pp of geographic
130	133	holdout $R^2$ relative to the market-status stage, and adding raw coordinates

modified paper/sections/titlepage.tex +1 −1

		`@@ -41,7 +41,7 @@`
41	41
42	42	% --- Date & Version ---
43	43	{\normalsize First draft: May 2026}\\[0.1cm]
44		−{\normalsize This version: \today}\\[0.1cm]
	44	+{\normalsize This version: August 5, 2026}\\[0.1cm]
45	45	{\small\itshape Version~\WPversion}
46	46
47	47	\end{center}
48	48