Add ETL process for pregnancy calibration targets and update documenion

Makefile

@@ -61,6 +61,7 @@ database:
 	python policyengine_us_data/db/etl_snap.py
 	python policyengine_us_data/db/etl_state_income_tax.py
 	python policyengine_us_data/db/etl_irs_soi.py
+	python policyengine_us_data/db/etl_pregnancy.py
 	python policyengine_us_data/db/validate_database.py
 
 database-refresh:

changelog.d/563.added

@@ -0,0 +1 @@
+Impute pregnancy in CPS microdata using CDC VSRR birth counts and Census ACS female population, with calibration targets per state.

policyengine_us_data/datasets/cps/cps.py

@@ -294,6 +294,26 @@ def add_takeup(self):
     imputed_risk = rng.random(n_persons) < wic_risk_rate_by_person
     data["is_wic_at_nutritional_risk"] = receives_wic | imputed_risk
 
+    # Pregnancy: stochastically assign is_pregnant to women 15-44
+    # using CDC/Census-derived state-level pregnancy rates.
+    # CPS does not ask about pregnancy; calibration will fine-tune.
+    from policyengine_us_data.db.etl_pregnancy import (
+        get_state_pregnancy_rates,
+    )
+
+    pregnancy_rates = get_state_pregnancy_rates()
+    national_rate = 0.041  # fallback
+    pregnancy_rate_by_person = np.array(
+        [pregnancy_rates.get(s, national_rate) for s in person_states]
+    )
+    ages = data["age"]
+    is_female = data["is_female"]
+    is_eligible = is_female & (ages >= 15) & (ages <= 44)
+    rng = seeded_rng("is_pregnant")
+    data["is_pregnant"] = is_eligible & (
+        rng.random(n_persons) < pregnancy_rate_by_person
+    )
+
     # Voluntary tax filing: some people file even when not required and not
     # seeking a refund. EITC take-up already captures refund-seeking behavior
     # (if you take up EITC, you file). This variable captures people who file

policyengine_us_data/db/DATABASE_GUIDE.md

@@ -32,7 +32,8 @@ make promote-database   # Copy DB + raw inputs to HuggingFace clone
 | 6 | `etl_snap.py` | USDA FNS + Census ACS | SNAP participation (admin state-level, survey district-level) |
 | 7 | `etl_state_income_tax.py` | No | State income tax collections (Census STC FY2023, hardcoded) |
 | 8 | `etl_irs_soi.py` | IRS | Tax variables, EITC by child count, AGI brackets, conditional strata |
-| 9 | `validate_database.py` | No | Checks all target variables exist in policyengine-us |
+| 9 | `etl_pregnancy.py` | CDC VSRR + Census ACS | Pregnancy prevalence by state (provisional birth counts) |
+| 10 | `validate_database.py` | No | Checks all target variables exist in policyengine-us |
 
 ### Raw Input Caching
 
@@ -146,6 +147,7 @@ Strata are categorized by their **constraints**, not by a separate group ID fiel
 | `adjusted_gross_income` | Income/AGI brackets |
 | `snap` | SNAP recipient strata |
 | `medicaid_enrolled` | Medicaid enrollment strata |
+| `is_pregnant` | Pregnancy prevalence strata |
 | `eitc_child_count` | EITC recipients by qualifying children |
 | `state_income_tax` | State-level income tax collections |
 | `aca_ptc` | ACA Premium Tax Credit strata |

policyengine_us_data/db/create_field_valid_values.py

@@ -74,6 +74,7 @@ def populate_field_valid_values(session: Session) -> None:
         ("source", "USDA FNS SNAP", "administrative"),
         ("source", "Census ACS S2201", "survey"),
         ("source", "Census STC", "administrative"),
+        ("source", "CDC VSRR Natality", "administrative"),
         ("source", "PolicyEngine", "hardcoded"),
     ]