SPB Git
spboucher.ai

Repository Public Research preview

spbwp3_uqo

Page 1 · 11 commits on main

  1. Wed, Nov 4, 20261 commit

    1. R03-ADVE-06/DATA-07/THEO-09/ECON-03/ECON-08/EDIT-08: quantile regressions on the corrected design (laptop, 3.4 h)
      ···
      - 5 tau fits on the trimmed v2 design (150,000-listing subsample), all converged; parking coefficient of ordinary
  magnitude (|beta| <= 0.03 vs 691/317/-98/-527 in v1.2); pseudo-R2 0.429/0.415/0.413/0.435/0.472
- legacy check: the v1.1 QR coefficients reproduce to 1e-5 (62 of 63); the stored 'pseudo-R2' row equals 1 - R1(tau)
- paired obs + ZIP3 bootstrap, 999 draws each, all converged: of 14 inter-quantile differences, 13 independent,
  10 normal-approx., 10 percentile CI, 8 Holm, 10 BH, 8 Romano-Wolf; cluster-bootstrap SEs for tau 0.10/0.90

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 10 files +265

  2. Tue, Nov 3, 20268 commits

    1. R03-ADVE-06: QR legacy check records that the stored pseudo-R2 equals 1 - R1(tau)
      ···
      Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 1 file +3
    2. R03-DATA-08: environment of node M4BP48 (lock, hardware); DAS days-on-Zillow IQR (raw vs sample)
      ···
      Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 3 files +6 −6
    3. R03-ADVE-08/ECON-02/THEO-06/ECON-04/THEO-05/EDIT-07/ADVE-14/ECON-07/EDIT-09/DATA-03/DATA-04: wild bootstrap, state regressions with SEs, full-sample Moran, Conley HAC; table generator, run_all.sh, test_tables
      ···
      - wild cluster bootstrap by state (B = 9,999): 23 of 63 slopes at 5% (Webb and Rademacher), 9 after Holm
- state-by-state OLS with ZIP3-cluster SEs: reversals with |t| > 2 are 0-12 per variable (v1.2: 19-23 by sign only);
  homogeneity rejected for every variable; EB shrinkage bands; fully interacted +6.1 pp over state FE (0.752 vs 0.690)
- Moran's I on the full random test set (157,769 listings, k = 8): OLS 0.454, LightGBM 0.288, XGBoost 0.234; distance
  bands 0-1 / 1-5 / 5-25 km; Conley HAC 25/50/100 km on 100,000 listings (between HC0 and state-cluster SEs)
- scripts/08_tables_figures_v2.py: every table as a v2_*.tex fragment + figures + inline_numbers.json (re-runnable as
  batches complete); run_all.sh entry point; tests/test_tables.py

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 14 files +1,757 −5
    4. R03-DATA-06/EDIT-04/ADVE-05/ADVE-12/ECON-10/THEO-08/EDIT-02/ECON-08/ADVE-17: inference v2 (fast steps) and geo-inference script
      ···
      - 06_inference_v2.py: zpid/split assertions pass; v2 OLS cluster SEs: 33 of 63 slopes |t|>1.96 under state
  clustering (t(50): 33; Holm 17; BH 26), CSS effective G (state: median 14.2, min 3.7); '000' cluster = 16
  listings, nearest-ZIP3 reassignment; clustered Kennedy; lot recoding on v2 (0.0182 -> 0.0958, matches
  v1.2 Table 29); paired squared-error differences with ZIP3-cluster SEs (stored predictions)
- 09_geo_inference.py: ZIP3-cluster bootstrap of the OLS-vs-ensemble gap per design, LOSO paired tests,
  SSE decomposition, feasible level correction, 2x2 region x rounds, duplicates re-score, gap range

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 12 files +1,077
    5. R03-EDIT-01: 05_geo_designs OLS/FE fits cast to float64 (float32 LinearRegression returns rank 1)
      ···
      Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 1 file +3 −1
    6. R03-DATA-04/ADVE-13/ECON-12/ECON-09: legacy exhibits machine-generated; OLS predictions via centred least squares
      ···
      - 04_export_tables.py --legacy-exhibits: Tables 2, 5, 9, 11, 17, 30 from stored artefacts / rebuilt sample
  (Table 9 reproduced: 0.634/0.630, 0.681/0.678, 0.729/0.725; Ridge/Lasso/ENet rows flagged not reproducible;
  Moran SD under both ddof conventions 0.0076 / 0.0094; z-statistics not stored)
- OLS out-of-fold predictions use LinearRegression (centred): a raw-constant lstsq on the legacy design with the
  corrupt parking column gives 0.613 instead of the stored 0.630 on the random split (now reproduced to 1e-12)

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 15 files +250 −10
    7. R03-EDIT-01/ADVE-01/ADVE-03/ADVE-04/EDIT-03/THEO-01/ADVE-06/ECON-03/ADVE-05/THEO-15: geo-design runner, QR v2 and stored-estimate checks
      ···
      - scripts/05_geo_designs.py: OLS/XGB/LGB under random, size-matched, ten-state, GroupKFold by state/ZIP3/ZIP5
  and LOSO on the v2 design; nested inner GroupKFold(3) round selection; 5 seeds; factorial ablation; OOF predictions;
  resumable (smoke-tested on a 15k/30k subsample on the laptop)
- scripts/07_qr_v2.py: QR at 5 tau on the trimmed v2 design with convergence flags, 10-seed stability, paired
  obs + ZIP3 bootstrap (999 draws), Holm/BH/Romano-Wolf, percentile/basic CIs, legacy check vs qr_results.pkl
- scripts/10_stored_checks.py: counts over 62 slopes (|t|>1.96: 32 state / 34 ZIP3; t(50): 31; Holm 19; BH 29),
  oracle within-state R2 on the stored random split (XGB 0.802, LGB 0.774, OLS 0.599), '000' cluster = 16,
  Moran SD conventions, LOSO summaries, 0.547/0.542 traced to extended_results.pkl, duplicates (1,127 test rows,
  R2 change < 0.001), Zestimate selection, clustered Kennedy; results/inline_numbers.json (82 entries)

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 15 files +1,106
    8. R03-DATA-01/EDIT-05/ADVE-07/ADVE-10/DATA-06/THEO-15/DATA-02/DATA-08: rebuild the upstream stage (00_build_sample, 00b_design_matrix, 00c --legacy, tests, DAS, environment)
      ···
      - scripts/00_build_sample.py + src/wp3/build.py: DuckDB -> analytical sample; all 68 columns equal
  analytical_sample.pkl bit-for-bit (max |d| = 0); attrition, descriptives, corrected Table 3 rules
  (stories/parking = state median), lot reconciliation (123,942 + 6,115 + 211 + 3 negative = 130,271),
  duplicates (6,787 rows / 3,152 groups), listing-type shares, data window (snapshot 2026-01-08)
- data/ref/category_map.csv: frozen raw-string -> category lookup (30,038 strings)
- scripts/00b_design_matrix.py + src/wp3/design.py: one design matrix (design_v2.parquet); legacy mode
  equals model_data.pkl::X_const (max |d| = 0, means/stds ddof=1); v2 = ln lot recoded + lot_missing,
  parking/stories winsorised, implausible flag (4,970), six missingness indicators
- scripts/00c_primary_models.py --legacy --ols-only: baseline OLS reproduced (params rel 7e-12, HC3 SE
  rel 9e-8, R2 0.634277, F 18,712.47, skew 0.1776, kurtosis 5.5452; state-FE test R2 0.67757 vs 0.67754)
- tests/: 13 assertions green; environment/: requirements.lock (laptop + M4M64c), README
- data/DATA_AVAILABILITY.md: source, window, licence, SHA-256, sizes, listing types

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 34 files +32,327 −1

  3. Sat, Oct 17, 20264 commits

    1. révision scientifique complète (editorial board, 2026-10-17) — référence non citée retirée
      ···
      Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 4 files +6 −17
    2. révision scientifique complète (editorial board, 2026-10-17)
      ···
      v1.1 → v1.2 (October 2026). Audit des données (variables scolaires imputées à 93 %, ln(lot)=0 pour 16,5 %,
valeurs aberrantes), inférence groupée ZIP3/État, estimation État par État, bootstrap apparié des tests
inter-quantiles, validation géographique étendue (courbe d'arbres, split apparié, GroupKFold, leave-one-state-out,
R² within-state), OLS hors État corrigé (0,539), 23 références vérifiées ajoutées + 28 DOI, sections réécrites,
docs/REVIEW-REPORT.md et docs/CHANGELOG-REVISION.md, scripts 05–08, 82 pages.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 60 files +4,379 −1,085
    3. PDF de référence avant révision scientifique
      ···
      Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 1 file
    4. harmonisation standard UQO WP (2026-10-17)
      ···
      Page titre du gabarit, natbib + aer, docs/ pour les rapports d'audit, README au standard (Status, citation WP 3), .gitignore et Makefile/.latexmkrc du gabarit, main.pdf recompilé.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
      Simon-Pierre Boucher 10 files +196 −414

  4. Wed, Aug 5, 20266 commits

    1. Record author validation of the novelty sentence (kept as written)
      ···
      All five flagged items in UPGRADE_REPORT are now resolved.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher 1 file +3 −3
    2. Record author decisions: Waltl cited as 2019 (print issue); repo made public
      ···
      Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher 1 file +3 −3
    3. 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 6 files +27 −11
    4. Final QA pass: verify flagged magnitudes, fix clipped tables, refresh README
      ···
      - Verified Black (1999) headline against QJE abstract: corrected 2% to
  2.5% per 5% test-score increase; verified Campbell et al. (2011) 27%
  foreclosure discount (text unchanged).
- Fixed two pre-existing table-clipping defects: Table 5 (OLS results)
  split across two pages via ContinuedFloat; appendix landscape table
  compacted. Build now has 0 overfull vboxes; no values changed.
- README badges updated (60 pages, 69 verified references, v1.1).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher 6 files +45 −15
    5. Scholarly upgrade of the paper (v1.1): literature expansion + rewrite
      ···
      - Bibliography: 37 -> 69 references, every new entry verified via
  OpenAlex (exact metadata + DOI). Removed one fabricated reference
  (chen2020housing) and corrected two mis-dated entries (Bourassa et
  al. 2010; Meyer & Pebesma 2021).
- Rewrote and expanded Introduction, Literature Review (10 thematic
  subsections), Discussion, and Conclusion; citation-grounded edits in
  Data, Methodology, Results, and Limitations.
- No results, tables, or figures changed. Compiles clean: 61 pages,
  0 undefined references, 69/69 entries cited.
- Added PAPER_REVIEW.md (critical assessment) and UPGRADE_REPORT.md
  (per-reference justification, flagged claims, contradicting
  literature).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher 13 files +1,217 −100
    6. Initial commit: WP3 research compendium (hedonic housing prices, US)
      ···
      Restructured research project for UQO Working Paper No. 3: refactored
analysis pipeline (src/wp3 + numbered scripts), 11 publication-ready
figures regenerated from stored artifacts, LaTeX paper (52 pages,
BibTeX bibliography, compiles clean), CSV table exports, and full
audit/change documentation (AUDIT.md, CHANGES.md).

Large data artifacts (DuckDB raw extract, processed pickles) are
excluded from the repository and available on request.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
      simon-pierre boucher 54 files +4,333
Menu

Theme
↑↓navigate ↵open escclose /⌘Kanywhere