Add local attribution factor builders

tests/test_factor_attribution.py

@@ -9,6 +9,7 @@ from pathlib import Path
 from urllib.error import URLError
 from unittest import mock
 
+import numpy as np
 import pandas as pd
 
 from factor_attribution import (
@@ -17,6 +18,8 @@ from factor_attribution import (
     KEN_FRENCH_DAILY_FF5_ZIP_URL,
     _download_kf_zip_bytes,
     _parse_kf_daily_csv,
+    build_extension_factors,
+    build_proxy_core_factors,
     load_external_us_factors,
 )
 
@@ -289,3 +292,46 @@ class ExternalFactorLoaderTests(unittest.TestCase):
         with zipfile.ZipFile(buffer, mode="w") as archive:
             archive.writestr(filename, contents)
         return buffer.getvalue()
+
+
+class LocalFactorConstructionTests(unittest.TestCase):
+    def test_build_extension_factors_returns_expected_columns_with_non_null_values_after_warmup(self):
+        prices = self._make_price_frame(benchmark="SPY")
+
+        factors = build_extension_factors(prices, benchmark="SPY", market="us")
+
+        self.assertListEqual(list(factors.columns), ["MOM", "LOWVOL", "RECOVERY"])
+        self.assertTrue(factors.iloc[260:].notna().all().all())
+        self.assertGreater(factors.iloc[260:].abs().sum().sum(), 0.0)
+
+    def test_build_proxy_core_factors_returns_expected_columns_with_non_null_values_after_warmup(self):
+        prices = self._make_price_frame(benchmark="000300.SS")
+
+        factors = build_proxy_core_factors(prices, benchmark="000300.SS", market="cn")
+
+        self.assertListEqual(
+            list(factors.columns),
+            ["MKT", "SMB_PROXY", "HML_PROXY", "RMW_PROXY", "CMA_PROXY"],
+        )
+        self.assertTrue(factors.iloc[260:].notna().all().all())
+        self.assertGreater(factors.iloc[260:].abs().sum().sum(), 0.0)
+
+    def _make_price_frame(self, benchmark: str) -> pd.DataFrame:
+        dates = pd.date_range("2025-01-01", periods=320, freq="B")
+        steps = np.arange(len(dates), dtype=float)
+        symbols = [
+            ("A", 45.0, 0.0006, 0.030, 19.0, 0.1),
+            ("B", 60.0, 0.0003, 0.025, 23.0, 0.8),
+            ("C", 75.0, -0.0002, 0.035, 17.0, 1.4),
+            ("D", 90.0, 0.0008, 0.020, 29.0, 0.5),
+            ("E", 55.0, -0.0001, 0.028, 31.0, 1.9),
+            ("F", 70.0, 0.0005, 0.032, 21.0, 2.5),
+        ]
+        data = {}
+        for symbol, base, drift, amplitude, frequency, phase in symbols:
+            log_path = drift * steps + amplitude * np.sin(steps / frequency + phase)
+            data[symbol] = base * np.exp(log_path)
+
+        benchmark_path = 0.0004 * steps + 0.018 * np.sin(steps / 27.0 + 0.3)
+        data[benchmark] = 250.0 * np.exp(benchmark_path)
+        return pd.DataFrame(data, index=dates)