CLN: Use ruff to improve code quality

ci/azure_template_posix.yml

@@ -98,6 +98,7 @@ jobs:
      black --check linearmodels
      isort --check linearmodels
      ruff check linearmodels
+     flake8 linearmodels
     displayName: 'Check style and formatting'
 
   - script: |

linearmodels/__init__.py

@@ -35,6 +35,7 @@
 from __future__ import annotations
 
 import os
+import sys
 
 from ._version import version as __version__, version_tuple
 from .asset_pricing.model import (
@@ -55,34 +56,34 @@
 from .system import IV3SLS, SUR, IVSystemGMM
 
 OLS = _OLS
-WARN_ON_MISSING = os.environ.get("LINEARMODELS_WARN_ON_MISSING", True)
-WARN_ON_MISSING = False if WARN_ON_MISSING in ("0", "False") else True
-DROP_MISSING = os.environ.get("LINEARMODELS_DROP_MISSING", True)
-DROP_MISSING = False if DROP_MISSING in ("0", "False") else True
+WARN_ON_MISSING = os.environ.get("LINEARMODELS_WARN_ON_MISSING", "1")
+WARN_ON_MISSING = False if WARN_ON_MISSING in ("", "0", "false", "False") else True
+DROP_MISSING = os.environ.get("LINEARMODELS_DROP_MISSING", "1")
+DROP_MISSING = False if DROP_MISSING in ("", "0", "false", "False") else True
 
 __all__ = [
-    "AbsorbingLS",
-    "PooledOLS",
-    "PanelOLS",
-    "FirstDifferenceOLS",
-    "BetweenOLS",
-    "RandomEffects",
-    "FamaMacBeth",
-    "IVLIML",
+    "DROP_MISSING",
+    "IV2SLS",
+    "IV3SLS",
     "IVGMM",
     "IVGMMCUE",
-    "IV2SLS",
+    "IVLIML",
     "OLS",
     "SUR",
-    "IV3SLS",
+    "WARN_ON_MISSING",
+    "AbsorbingLS",
+    "BetweenOLS",
+    "FamaMacBeth",
+    "FirstDifferenceOLS",
     "IVSystemGMM",
     "LinearFactorModel",
     "LinearFactorModelGMM",
+    "PanelOLS",
+    "PooledOLS",
+    "RandomEffects",
     "TradedFactorModel",
-    "WARN_ON_MISSING",
-    "DROP_MISSING",
-    "version_tuple",
     "__version__",
+    "version_tuple",
 ]
 
 
@@ -92,12 +93,11 @@ def test(
     append: bool = True,
     location: str = "",
 ) -> int:
-    import sys
 
     try:
-        import pytest
-    except ImportError:  # pragma: no cover
-        raise ImportError("Need pytest to run tests")
+        import pytest  # noqa: PLC0415
+    except ImportError as exc:  # pragma: no cover
+        raise ImportError("Need pytest to run tests") from exc
 
     cmd = ["--tb=auto"]
     if extra_args:
@@ -117,7 +117,7 @@ def test(
         print(pkg)
     if not os.path.exists(pkg):
         raise RuntimeError(f"{pkg} was not found. Unable to run tests")
-    cmd = [pkg] + cmd
+    cmd = [pkg, *cmd]
     print("running: pytest {}".format(" ".join(cmd)))
     status = pytest.main(cmd)
     if exit:  # pragma: no cover

linearmodels/asset_pricing/__init__.py

@@ -1,3 +1,3 @@
 from .model import LinearFactorModel, LinearFactorModelGMM, TradedFactorModel
 
-__all__ = ["TradedFactorModel", "LinearFactorModelGMM", "LinearFactorModel"]
+__all__ = ["LinearFactorModel", "LinearFactorModelGMM", "TradedFactorModel"]

linearmodels/asset_pricing/covariance.py

@@ -4,8 +4,7 @@
 
 from __future__ import annotations
 
-import numpy
-from numpy import empty
+from numpy import empty, ndarray
 from numpy.linalg import inv
 
 from linearmodels.iv.covariance import (
@@ -20,7 +19,7 @@ class _HACMixin:
     def __init__(self, kernel: str, bandwidth: float | None) -> None:
         self._kernel: str | None = None
         self._bandwidth: float | None = None  # pragma: no cover
-        self._moments: numpy.ndarray = empty((0,))  # pragma: no cover
+        self._moments: ndarray = empty((0,))  # pragma: no cover
         self._check_kernel(kernel)
         self._check_bandwidth(bandwidth)
 
@@ -56,8 +55,8 @@ def _check_bandwidth(self, bandwidth: float | None) -> None:
             try:
                 assert bandwidth is not None
                 bandwidth = float(bandwidth)
-            except (TypeError, ValueError):
-                raise TypeError("bandwidth must be either None or a float")
+            except (TypeError, ValueError) as exc:
+                raise TypeError("bandwidth must be either None or a float") from exc
             if bandwidth < 0:
                 raise ValueError("bandwidth must be non-negative.")
 
@@ -101,8 +100,8 @@ def __init__(
         self,
         xe: linearmodels.typing.data.Float64Array,
         *,
-        jacobian: numpy.ndarray | None = None,
-        inv_jacobian: numpy.ndarray | None = None,
+        jacobian: ndarray | None = None,
+        inv_jacobian: ndarray | None = None,
         center: bool = True,
         debiased: bool = False,
         df: int = 0,
@@ -234,8 +233,8 @@ def __init__(
         self,
         xe: linearmodels.typing.data.Float64Array,
         *,
-        jacobian: numpy.ndarray | None = None,
-        inv_jacobian: numpy.ndarray | None = None,
+        jacobian: ndarray | None = None,
+        inv_jacobian: ndarray | None = None,
         kernel: str | None = None,
         bandwidth: float | None = None,
         center: bool = True,

linearmodels/asset_pricing/model.py

@@ -11,7 +11,6 @@
 from formulaic.materializers.types import NAAction
 import numpy as np
 from numpy.linalg import lstsq
-import pandas
 from pandas import DataFrame
 from scipy.optimize import minimize
 
@@ -99,7 +98,7 @@ def __repr__(self) -> str:
     def _drop_missing(self) -> linearmodels.typing.data.BoolArray:
         data = (self.portfolios, self.factors)
         missing = cast(
-            linearmodels.typing.data.BoolArray,
+            "linearmodels.typing.data.BoolArray",
             np.any(np.c_[[dh.isnull for dh in data]], 0),
         )
         if any(missing):
@@ -123,11 +122,11 @@ def _validate_data(self) -> None:
             )
         self._drop_missing()
 
-        p = cast(linearmodels.typing.data.Float64Array, self.portfolios.ndarray)
-        f = cast(linearmodels.typing.data.Float64Array, self.factors.ndarray)
+        p = cast("linearmodels.typing.data.Float64Array", self.portfolios.ndarray)
+        f = cast("linearmodels.typing.data.Float64Array", self.factors.ndarray)
         if has_constant(p)[0]:
             raise ValueError(
-                "portfolios must not contains a constant or "
+                "portfolios must not contain a constant or "
                 "equivalent and must not have rank\n"
                 "less than the dimension of the smaller shape."
             )
@@ -153,7 +152,7 @@ def formula(self, value: str | None) -> None:
 
     @staticmethod
     def _prepare_data_from_formula(
-        formula: str, data: pandas.DataFrame, portfolios: pandas.DataFrame | None
+        formula: str, data: DataFrame, portfolios: DataFrame | None
     ) -> tuple[DataFrame, DataFrame, str]:
         orig_formula = formula
         na_action = NAAction("raise")
@@ -224,9 +223,9 @@ def __init__(self, portfolios: IVDataLike, factors: IVDataLike):
     def from_formula(
         cls,
         formula: str,
-        data: pandas.DataFrame,
+        data: DataFrame,
         *,
-        portfolios: pandas.DataFrame | None = None,
+        portfolios: DataFrame | None = None,
     ) -> TradedFactorModel:
         """
         Parameters
@@ -349,7 +348,7 @@ def fit(
                 kernel=kernel,
             )
             bw = cov_est.bandwidth
-            _cov_config = {k: v for k, v in cov_config.items()}
+            _cov_config = dict(cov_config.items())
             _cov_config["bandwidth"] = bw
             rp_cov_est = KernelCovariance(
                 fe,
@@ -388,10 +387,10 @@ def fit(
         param_names = []
         for portfolio in self.portfolios.cols:
             param_names.append(f"alpha-{portfolio}")
-            for factor in self.factors.cols:
-                param_names.append(f"beta-{portfolio}-{factor}")
-        for factor in self.factors.cols:
-            param_names.append(f"lambda-{factor}")
+            param_names.extend(
+                [f"beta-{portfolio}-{factor}" for factor in self.factors.cols]
+            )
+        param_names.extend([f"lambda-{factor}" for factor in self.factors.cols])
 
         res = AttrDict(
             params=params,
@@ -540,9 +539,9 @@ def __init__(
     def from_formula(
         cls,
         formula: str,
-        data: pandas.DataFrame,
+        data: DataFrame,
         *,
-        portfolios: pandas.DataFrame | None = None,
+        portfolios: DataFrame | None = None,
         risk_free: bool = False,
         sigma: linearmodels.typing.data.ArrayLike | None = None,
     ) -> LinearFactorModel:
@@ -693,12 +692,12 @@ def fit(
         param_names = []
         for portfolio in self.portfolios.cols:
             param_names.append(f"alpha-{portfolio}")
-            for factor in self.factors.cols:
-                param_names.append(f"beta-{portfolio}-{factor}")
+            param_names.extend(
+                [f"beta-{portfolio}-{factor}" for factor in self.factors.cols]
+            )
         if not excess_returns:
             param_names.append("lambda-risk_free")
-        for factor in self.factors.cols:
-            param_names.append(f"lambda-{factor}")
+        param_names.extend([f"lambda-{factor}" for factor in self.factors.cols])
 
         # Pivot vcv to remove unnecessary and have correct order
         order = np.reshape(np.arange(s1), (nport, nf + 1))
@@ -777,7 +776,7 @@ def _moments(
         sigma_inv = self._sigma_inv
 
         f = self.factors.ndarray
-        nobs, nf, nport, _, s1, s2, s3 = self._boundaries()
+        nobs, nf, nport, _, _, _, _ = self._boundaries()
         fc = np.c_[np.ones((nobs, 1)), f]
         f_rep = np.tile(fc, (1, nport))
         eps_rep = np.tile(eps, (nf + 1, 1))
@@ -844,9 +843,9 @@ def __init__(
     def from_formula(
         cls,
         formula: str,
-        data: pandas.DataFrame,
+        data: DataFrame,
         *,
-        portfolios: pandas.DataFrame | None = None,
+        portfolios: DataFrame | None = None,
         risk_free: bool = False,
     ) -> LinearFactorModelGMM:
         """
@@ -1098,8 +1097,9 @@ def fit(
         r2 = 1.0 - residual_ss / total_ss
         param_names = []
         for portfolio in self.portfolios.cols:
-            for factor in self.factors.cols:
-                param_names.append(f"beta-{portfolio}-{factor}")
+            param_names.extend(
+                [f"beta-{portfolio}-{factor}" for factor in self.factors.cols]
+            )
         if not excess_returns:
             param_names.append("lambda-risk_free")
         param_names.extend([f"lambda-{f}" for f in self.factors.cols])

linearmodels/asset_pricing/results.py

@@ -338,7 +338,7 @@ def __init__(self, results: AttrDict):
     @property
     def std_errors(self) -> pd.DataFrame:
         """Estimated parameter standard errors"""
-        se = cast(linearmodels.typing.data.Float64Array, np.sqrt(np.diag(self._cov)))
+        se = cast("linearmodels.typing.data.Float64Array", np.sqrt(np.diag(self._cov)))
         ase = np.sqrt(np.diag(self._alpha_vcv))
         nportfolio, nfactor = self._params.shape
         nloadings = nportfolio * (nfactor - 1)

linearmodels/conftest.py

@@ -6,7 +6,7 @@
 
 logger = logging.getLogger(__name__)
 
-cow = bool(os.environ.get("LM_TEST_COPY_ON_WRITE", False))
+cow = bool(os.environ.get("LM_TEST_COPY_ON_WRITE", ""))
 if cow:
     try:
         pd.options.mode.copy_on_write = cow

linearmodels/datasets/birthweight/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame
 
+from linearmodels import datasets
+
 DESCR = """
 J. Mullahy (1997), "Instrumental-Variable Estimation of Count Data Models:
 Applications to Models of Cigarette Smoking Behavior," Review of Economics
@@ -23,6 +25,5 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     return datasets.load(__file__, "birthweight.csv.bz2")

linearmodels/datasets/card/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame
 
+from linearmodels import datasets
+
 DESCR = """
 D. Card (1995), "Using Geographic Variation in College Proximity to Estimate
 the Return to Schooling," in Aspects of Labour Market Behavior:  Essays in
@@ -44,6 +46,5 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     return datasets.load(__file__, "card.csv.bz2")

linearmodels/datasets/fertility/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame
 
+from linearmodels import datasets
+
 DESCR = """
 W. Sander, "The Effect of Women's Schooling on Fertility," Economics Letters
 40, 229-233.
@@ -35,6 +37,5 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     return datasets.load(__file__, "fertility.csv.bz2")

linearmodels/datasets/french/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame, to_datetime
 
+from linearmodels import datasets
+
 DESCR = """
 Data from Ken French's data library
 http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html
@@ -44,7 +46,6 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     data = datasets.load(__file__, "french.csv.bz2")
     data["dates"] = to_datetime(data.dates)

linearmodels/datasets/fringe/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame
 
+from linearmodels import datasets
+
 # noinspection SpellCheckingInspection,SpellCheckingInspection
 DESCR = """
 F. Vella (1993), "A Simple Estimator for Simultaneous Models with Censored
@@ -48,6 +50,5 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     return datasets.load(__file__, "fringe.csv.bz2")

linearmodels/datasets/jobtraining/__init__.py

@@ -1,5 +1,7 @@
 from pandas import DataFrame
 
+from linearmodels import datasets
+
 DESCR = """
 H. Holzer, R. Block, M. Cheatham, and J. Knott (1993), "Are Training Subsidies
 Effective? The Michigan Experience," Industrial and Labor Relations Review 46,
@@ -39,6 +41,5 @@
 
 
 def load() -> DataFrame:
-    from linearmodels import datasets
 
     return datasets.load(__file__, "jobtraining.csv.bz2")