# Contract (P0) — F6-dev-01

**Question.** Does the cross-country Preston curve (life expectancy vs. log real income per capita) hold in the most recent usable WDI year, and how strong is it? Descriptive/associational only.

**Estimand.** β in the linear projection LE_i = α + β·ln(GDPpc_i) + ε_i, one equally weighted obs per country/economy in the analysis year; LE = life expectancy at birth (years), GDPpc = GDP per capita PPP constant 2021 intl $. β = years of LE per log-point of GDPpc. Meaningful as a best-linear-predictor slope (no causal interpretation); requires finite variance and enough countries; robust (HC) SE for heteroskedasticity, assuming independence across countries.

**Required outputs.** out/results.json (analysis_year, n_excluded_missing, baseline_metric, model_metric, pearson_r, r_squared, effect_estimate, intercept, effect_se, effect_ci_low, effect_ci_high, conclusion, task_id, feasibility, feasibility_reason, data_issues, n_input_rows, n_analysed, random_seed); out/figure1.png (scatter + OLS line); out/table1.csv (income_level, n_countries, mean_log_gdp_pc, mean_life_expectancy); out/report.md (6 sections in order); out/analysis.py; out/claims.json; plus state files, exclusions.csv.

**Information rules.** Exclude entities with region == "Aggregates" in wb_entities.csv. Analysis year = most recent year 2015–2023 with both indicators observed for >=150 countries/economies. In that year drop countries with any missing indicator (count in n_excluded_missing), no imputation. Natural log. Only the listed files.

**Invalidity / impossibility.** No year with >=150 complete countries → infeasible (premise_invalid). Non-positive GDP (log undefined). Entities in indicators not matched to metadata (ambiguous country status) must be handled explicitly. Using aggregates would invalidate the estimate.