Declare the NZ transport and entitlement graphs

changelog.d/nz-transport-graph.added.md

@@ -0,0 +1 @@
+Declare NZ transport and entitlement branches with independent calibration variants and an identity-factor hold-fixed sensitivity. Add a synthetic replay fixture, exact export and gate contracts, and positive-payment comparator inputs. A `money.minor_units@1` node rounds each weekly benefit net half-even to whole cents before its gross-up, and `geography.person_location@1`, in its own module, carries assigned household region codes to persons. The reg 17 base rate runs as one engine counterfactual and the reg 18 cutout as one zero-solve per population. The graph gates the Accommodation Supplement payment on the engine's non-rate judgment, excludes and counts superannuitant units outside S5, and pays them in S5 from NZ Superannuation before tax. Keep the package inactive while engine pins and money quantile producers remain open.

docs/transport-entitlement.md

@@ -177,16 +177,33 @@ comparing it with the hold-out reference. The toy graph
 does this with its invented family-credit engine. It supplies no real WFF
 inputs or statutory claim.
 
-The committed `build/nz/as_rate_bridge.json` sole-parent declaration rules out
-a `solve_zero` recipe for golden-08: no engine input combination reproduces
-the harness's single-with-children gross rate paired with the Income Test 3
-slope. It instead declares three evaluations, `gross_rate`,
-`reduction_at_anchor` and `reduction_after_step`, whose outputs recover the
-slope and lower threshold before assembling the cutout. The resource declares
-the overrides, anchor, step, composition and refusal conditions. The driver
-executes declared kernel calls but does not assemble this nonlinear
-composition. The approved executable golden-08 case and evaluation assembly
-therefore remain unresolved; no golden-08 result is claimed.
+`build/nz/as_rate_bridge.json` declares both NZ bridge outputs as kernel
+calls, with no new kernel. On each population that runs the Accommodation
+Supplement engine, a `calibration.prepare@1` node writes the family masks the
+equality-only bridge masks compare, a `simulate.counterfactual@1` node writes
+the reg 18(3)(b) sole-parent Jobseeker Support rate, and a
+`simulate.solve_zero@1` node writes the reg 18 cutout: the zero of each
+unit's Jobseeker Support residual over weekly income on [0, 5000]. A sole
+parent's residual is the sole-parent rate minus the Income Test 3 reduction;
+every other unit's residual is its own net payment. The resource keeps the
+harness expressions as the statement of equivalence. Each branch runs its own
+reg 17 base-rate counterfactual: zero-income Jobseeker Support once for a
+single unit and twice for a partnered one, plus the eldest-child family tax
+credit over 52 for a unit with a dependent child. The tests in
+`test_nz_as_bridge.py` reproduce golden-08's base rate and cutout with a
+labelled fixture engine; no real-engine result is claimed.
+
+rulespec-nz's rounded Accommodation Supplement payment never reads the
+module's non-rate entitlement judgment, so the NZ graph gates the payment in
+the host. Each branch's `positive_payment` node (`calibration.prepare@1`)
+holds where the payment exceeds zero and the judgment reads "holds" (code 1;
+undetermined fails), and, in every branch but S5, where the unit's receipt
+group is not superannuitant (MC11's exclude-and-count). A
+`simulate.counterfactual@1` node then writes
+`accommodation_supplement_entitled_weekly_payment`, the rounded payment
+masked on that indicator, which the dollar comparator, the reporting cells
+and the gates read. S5 instead pays superannuitant units from a reg 17 base
+rate that takes NZ Superannuation before tax.
 
 The hub can compare approved small cases and sampled-unit outputs without donor
 H5 I/O using the declared case format in

packages/microcosm-build/src/microcosm/build/calibration_predicates.py

@@ -81,6 +81,8 @@ def predicate_mask(
             part = values.eq(value)
         elif operator == "ge":
             part = values.ge(value)
+        elif operator == "gt":
+            part = values.gt(value)
         elif operator == "lt":
             part = values.lt(value)
         elif operator == "in" and isinstance(value, (list, tuple)) and value:
@@ -165,6 +167,8 @@ def _validate_input(values: pd.Series, domain: Mapping[str, Any], name: str) ->
             _finite_number(value) and value >= 0 and value == int(value)
             for value in values.tolist()
         )
+    elif kind == "nonnegative_number":
+        valid = all(_finite_number(value) and value >= 0 for value in values.tolist())
     else:
         allowed = domain["values"]
         valid = all(
@@ -178,7 +182,12 @@ def _validate_input(values: pd.Series, domain: Mapping[str, Any], name: str) ->
 def _input_domain_kind(domain: Mapping[str, Any], name: str) -> str:
     """Share domain-schema validation between activation and table validation."""
     kind = domain.get("kind")
-    if kind in ("identity", "signed_integer", "nonnegative_integer"):
+    if kind in (
+        "identity",
+        "signed_integer",
+        "nonnegative_integer",
+        "nonnegative_number",
+    ):
         return kind
     if isinstance(domain.get("values"), (list, tuple)) and domain["values"]:
         return "values"

packages/microcosm-build/src/microcosm/build/nz/as_encoding.json

@@ -0,0 +1,6 @@
+{
+  "rent_allocation": "reference_unit",
+  "asset_test": true,
+  "area_column": "as_area",
+  "rent_factor": 1
+}