Merge pull request #169 from PolicyEngine/nikhilwoodruff/issue168

Create and upload datasets for the new multi-year format

Author: nikhilwoodruff

Makefile

@@ -36,6 +36,7 @@ data:
 	python policyengine_uk_data/datasets/frs/enhanced_frs.py
 	python policyengine_uk_data/datasets/frs/local_areas/constituencies/calibrate.py
 	python policyengine_uk_data/datasets/frs/local_areas/local_authorities/calibrate.py
+	python policyengine_uk_data/utils/create_multi_year_dataset.py
 
 efrs:
 	python policyengine_uk_data/datasets/frs/enhanced_frs.py

changelog_entry.yaml

@@ -0,0 +1,4 @@
+- bump: minor
+  changes:
+    added:
+    - New multi-year dataset format for FRS and Enhanced FRS.

policyengine_uk_data/storage/upload_completed_datasets.py

@@ -7,6 +7,8 @@ def upload_datasets():
     dataset_files = [
         FRS_2023_24.file_path,
         EnhancedFRS_2023_24.file_path,
+        STORAGE_FOLDER / "frs_2023_29.h5",
+        STORAGE_FOLDER / "enhanced_frs_2023_29.h5",
         STORAGE_FOLDER / "parliamentary_constituency_weights.h5",
         STORAGE_FOLDER / "local_authority_weights.h5",
     ]

policyengine_uk_data/utils/create_multi_year_dataset.py

@@ -0,0 +1,53 @@
+from policyengine_uk.data import UKMultiYearDataset, UKSingleYearDataset
+from policyengine_uk.data.economic_assumptions import apply_uprating
+from policyengine_uk import Microsimulation
+from policyengine_uk_data.storage import STORAGE_FOLDER
+from policyengine_core.data import Dataset
+
+
+def convert_legacy_to_multi_year_dataset(
+    file_path: str,
+    new_file_path: str,
+    start_year: int = 2023,
+    end_year: int = 2029,
+) -> UKMultiYearDataset:
+    """
+    Convert a legacy single year dataset to a multi-year dataset.
+    """
+    sim = Microsimulation(dataset=Dataset.from_file(file_path))
+
+    dataset = UKSingleYearDataset.from_simulation(sim, fiscal_year=start_year)
+    dataset.time_period = str(start_year)
+
+    datasets = [dataset]
+
+    for year in range(start_year + 1, end_year + 1):
+        dataset = dataset.copy()
+        dataset.time_period = str(year)
+
+    multi_year_dataset = UKMultiYearDataset(
+        datasets=datasets,
+    )
+    multi_year_dataset = apply_uprating(multi_year_dataset)
+
+    multi_year_dataset.save(new_file_path)
+
+
+if __name__ == "__main__":
+    file_paths = [
+        STORAGE_FOLDER / "frs_2023_24.h5",
+        STORAGE_FOLDER / "enhanced_frs_2023_24.h5",
+    ]
+    out_file_paths = [
+        STORAGE_FOLDER / "frs_2023_29.h5",
+        STORAGE_FOLDER / "enhanced_frs_2023_29.h5",
+    ]
+
+    for file_path, new_file_path in zip(file_paths, out_file_paths):
+        convert_legacy_to_multi_year_dataset(
+            file_path=str(file_path),
+            new_file_path=str(new_file_path),
+            start_year=2023,
+            end_year=2029,
+        )
+        print(f"Converted {file_path} to {new_file_path}")