# Paper-to-Code Mapping # 논문 32개 모듈과 코드베이스 연동 맵 ## Module #14-16: Data Overview (데이터 개관) ### #14: Context - Quantum Computing Industry **논문 내용:** 양자컴퓨팅 산업 배경 설명 **코드 위치:** - `src/features.py::consolidate_company_snapshots()` - Filter: `df[df.sector_fe == 'quantum']` **테스트:** ```python # test/unit/test_features.py def test_quantum_sector_filtering(): """Verify quantum sector companies are correctly identified""" ``` ### #15: Sample Construction **논문 내용:** 표본 구성 (출처/코호트/필터) **코드 위치:** - `src/features.py::consolidate_company_snapshots()` - `src/features.py::engineer_features()` **생성 아티팩트:** 🗄️ 표본구성표 **테스트:** ```python # test/integration/test_sample_construction.py def test_sample_size_matches_paper(): """Verify final sample size matches paper Table X""" df = consolidate_company_snapshots('data/raw') assert len(df) == PAPER_REPORTED_N ``` ### #16: Variables Overview (E/L/V/F) **논문 내용:** - E: Early event (Series A at baseline) - L: Later success (Series B+ at endpoint) - V: Vagueness score - F: Flexibility (1 - is_hardware) **코드 위치:** - `src/features.py::engineer_features()` - `src/vagueness_v2.py::StrategicVaguenessScorerV2` **생성 아티팩트:** 🗄️ 변수정의·요약통계 **테스트:** ```python # test/integration/test_variables.py def test_table_descriptive_stats(): """Generate Table X: Descriptive Statistics and compare with paper""" df = load_and_engineer_features() stats = df[['E', 'L', 'V', 'F']].describe() # Compare with paper-reported values ``` --- ## Module #17-22: Empirical Methodology (실증 방법론) ### #17: Measurements **논문 내용:** V (vagueness), F (flexibility) 측정 방법 **코드 위치:** - `src/vagueness_v2.py::StrategicVaguenessScorerV2.score()` - `src/features.py::engineer_features()` (F = 1 - is_hardware) **테스트:** ```python # test/unit/test_vagueness.py def test_vagueness_measurement_validity(): """Validate vagueness scoring against hand-coded examples""" ``` ### #19: Identification Strategy **논문 내용:** 내생성 논의, 식별 전략 **코드 위치:** - `src/models.py` (주석 참조) - Controls selection logic **문서화:** `docs/identification_strategy.md` ### #20: Main Specifications **논문 내용:** E ~ V (OLS), L ~ V × F (Logit) **코드 위치:** - `src/models.py::run_HEV()` - H1 for E ~ V - `src/models.py::run_HLVF()` - H2 for L ~ V × F **생성 아티팩트:** 🗄️ T_MainSpecs **테스트:** ```python # test/integration/test_main_specs.py def test_h1_specification_matches_paper(): """Verify H1 formula matches paper equation (X)""" # Run model result = run_HEV(df) # Check formula includes correct controls assert 'z_V' in result.params ``` --- ## Module #23-27: Results (결과) ### #23: H1 - Early Funding Penalty **논문 내용:** Vagueness reduces early funding **코드 위치:** - `src/models.py::test_h1_early_funding()` - OR `src/models.py::run_HEV()` (two-snapshot mode) **생성 아티팩트:** 🖼️ Fig2_EVF, 🗄️ T1 **테스트:** ```python # test/integration/test_paper_results.py def test_table1_h1_results(): """Reproduce Table 1: H1 regression coefficients""" df = load_analysis_data() result = test_h1_early_funding(df) # Compare with paper-reported values (with tolerance) PAPER_COEF_VAGUENESS = -0.234 # From paper Table 1 PAPER_SE = 0.089 assert abs(result.params['z_vagueness'] - PAPER_COEF_VAGUENESS) < 0.01 assert abs(result.bse['z_vagueness'] - PAPER_SE) < 0.01 def test_figure2_evf_plot(): """Reproduce Figure 2: E-V-F relationship""" df = load_analysis_data() fig_path = plot_figure2_evf(df) # Check file exists and is non-trivial assert Path(fig_path).exists() assert Path(fig_path).stat().st_size > 10000 ``` ### #24: H2 - Later Success Benefit **논文 内容:** Vagueness beneficial for later success (moderated by F) **코드 위치:** - `src/models.py::test_h2_main_growth()` - OR `src/models.py::run_HLVF()` (two-snapshot mode) **생성 아티팩트:** 🖼️ Fig3_LVF, Fig4_STV, 🗄️ T2 **테스트:** ```python def test_table2_h2_results(): """Reproduce Table 2: H2 logit regression""" df = load_analysis_data() result = test_h2_main_growth(df) # Main effect (vagueness) PAPER_COEF_MAIN = 0.456 assert abs(result.params['z_vagueness'] - PAPER_COEF_MAIN) < 0.01 # Interaction (vagueness × hardware) PAPER_COEF_INTERACTION = -0.321 interaction_param = [p for p in result.params.index if 'z_vagueness' in p and 'is_hardware' in p][0] assert abs(result.params[interaction_param] - PAPER_COEF_INTERACTION) < 0.01 ``` ### #25: H2a - V×F Interaction **논문 내용:** 상호작용 효과 (flexibility amplifies vagueness benefit) **코드 위치:** Same as #24 **생성 아티팩트:** 🖼️ Fig3_LVF (interaction plot) **테스트:** ```python def test_figure3_interaction_plot(): """Reproduce Figure 3: V×F interaction visualization""" df = load_analysis_data() result = test_h2_main_growth(df) # Generate interaction plot fig_path = plot_figure3_interaction(df, result) # Visual regression test (compare with reference image) # assert image_similarity(fig_path, 'test/fixtures/fig3_reference.png') > 0.95 ``` ### #26: Mechanisms - Pivot/Learning **논문 내용:** 메커니즘 분석 (피벗 빈도, 학습 속도) **코드 위치:** - ⚠️ **새로 작성 필요:** `src/models.py::test_mechanism_pivot()` **생성 아티팩트:** 🗄️ T_Mech **테스트:** ```python # test/unit/test_mechanisms.py def test_pivot_frequency_analysis(): """Mechanism: Companies with higher V pivot more frequently""" # TODO: Implement pivot detection logic # df['pivot_count'] = detect_pivots(df) # model = smf.ols('pivot_count ~ z_vagueness + controls', df).fit() # assert model.params['z_vagueness'] > 0 ``` ### #27: Robustness - Spec Curve **논문 내용:** 강건성 검증 (스펙 커브, 안정성) **코드 위치:** - `src/multiverse.py` (already exists) **생성 아티팩트:** 🗄️ T_SpecCurve **테스트:** ```python # test/integration/test_robustness.py def test_specification_curve(): """Run multiverse analysis across 100+ specifications""" from multiverse import run_specification_curve results = run_specification_curve(df) # Check majority of specs support main hypothesis significant_positive = sum((r.params['z_vagueness'] > 0) & (r.pvalues['z_vagueness'] < 0.05) for r in results) assert significant_positive / len(results) > 0.8 # 80%+ support ``` --- ## Module #11-13: Conceptual Model (개념틀) ### #11: Framework (2×2) **논문 내용:** 2×2 framework (시간×레벨) **생성 아티팩트:** 🖼️ Fig1_LV (conceptual figure) **코드 위치:** Manual illustration (not code-generated) **문서화:** `docs/conceptual_framework.md` ### #13: Hypotheses (H1/H2/H2a) **논문 내용:** 가설 정식화 **코드 위치:** - Documented in `src/models.py` docstrings - Each hypothesis test function has detailed docstring **테스트:** ```python # test/meta/test_hypothesis_documentation.py def test_h1_docstring_matches_paper(): """Verify H1 docstring matches paper hypothesis statement""" import inspect from models import test_h1_early_funding docstring = inspect.getdoc(test_h1_early_funding) assert "vagueness reduces early funding" in docstring.lower() ``` --- ## Summary Table: Code-Paper Mapping | Module | 논문 섹션 | 코드 파일 | 테스트 파일 | 아티팩트 | 상태 | |--------|----------|----------|------------|---------|-----| | #14 | Context | features.py | test_features.py | - | ✅ | | #15 | Sample | features.py | test_sample_construction.py | 🗄️표본구성 | ⚠️ | | #16 | Variables | features.py, vagueness_v2.py | test_variables.py | 🗄️요약통계 | ⚠️ | | #17 | Measurements | vagueness_v2.py | test_vagueness.py | - | ✅ | | #20 | Specifications | models.py | test_main_specs.py | 🗄️MainSpecs | ⚠️ | | #23 | H1 Results | models.py::test_h1 | test_paper_results.py | 🖼️Fig2, 🗄️T1 | ⚠️ | | #24 | H2 Results | models.py::test_h2 | test_paper_results.py | 🖼️Fig3-4, 🗄️T2 | ⚠️ | | #25 | Interaction | models.py::test_h2 | test_paper_results.py | 🖼️Fig3 | ⚠️ | | #26 | Mechanisms | models.py (new) | test_mechanisms.py | 🗄️T_Mech | ❌ | | #27 | Robustness | multiverse.py | test_robustness.py | 🗄️SpecCurve | ❌ | Legend: - ✅ = 코드 & 테스트 완료 - ⚠️ = 코드 있음, 테스트 필요 - ❌ = 코드 or 테스트 모두 필요