AnomalyLab 0.4.2__tar.gz → 0.5.0__tar.gz

{anomalylab-0.4.2 → anomalylab-0.5.0/AnomalyLab.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: AnomalyLab
-Version: 0.4.2
+Version: 0.5.0
 Summary: A Python package for empirical asset pricing analysis.
 Author: FinPhd
 Author-email: chenhaiwei@stu.sufe.edu.cn

{anomalylab-0.4.2 → anomalylab-0.5.0}/AnomalyLab.egg-info/SOURCES.txt

@@ -12,7 +12,6 @@ AnomalyLab.egg-info/dependency_links.txt
 AnomalyLab.egg-info/requires.txt
 AnomalyLab.egg-info/top_level.txt
 anomalylab/__init__.py
-anomalylab/config.py
 anomalylab/core/__init__.py
 anomalylab/core/core.py
 anomalylab/datasets/__init__.py
@@ -38,7 +37,6 @@ anomalylab/structure/data.py
 anomalylab/structure/panel_data.py
 anomalylab/structure/time_series.py
 anomalylab/utils/__init__.py
-anomalylab/utils/imports.py
 anomalylab/utils/utils.py
 anomalylab/visualization/__init__.py
 anomalylab/visualization/format.py

{anomalylab-0.4.2/AnomalyLab.egg-info → anomalylab-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: AnomalyLab
-Version: 0.4.2
+Version: 0.5.0
 Summary: A Python package for empirical asset pricing analysis.
 Author: FinPhd
 Author-email: chenhaiwei@stu.sufe.edu.cn

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/__init__.py

@@ -11,8 +11,7 @@ from anomalylab.empirical import (
 )
 from anomalylab.preprocess import FillNa, Normalize, OutlierHandler, Shift
 from anomalylab.structure import PanelData, TimeSeries
-from anomalylab.utils import *
-from anomalylab.utils.imports import *
+from anomalylab.utils import pp
 from anomalylab.visualization import FormatExcel
 
 __all__: list[str] = [

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/core/core.py

@@ -1,8 +1,12 @@
 from __future__ import annotations
 
-from importlib import resources
+from dataclasses import dataclass, field
+from importlib import resources  # noqa: F401
+from typing import Literal, Optional, Union
+
+import pandas as pd
+from pandas import DataFrame
 
-from anomalylab.config import *
 from anomalylab.empirical import (
     Correlation,
     FamaMacBethRegression,
@@ -12,8 +16,7 @@ from anomalylab.empirical import (
 )
 from anomalylab.preprocess import FillNa, Normalize, OutlierHandler, Shift
 from anomalylab.structure import PanelData, TimeSeries
-from anomalylab.utils import *
-from anomalylab.utils.imports import *
+from anomalylab.utils import Columns, Scalar, pp
 from anomalylab.visualization import FormatExcel
 
 
@@ -126,8 +129,7 @@ class Panel:
         return self._fm_preprocessor
 
     def format_preprocessor(self, path: str) -> FormatExcel:
-        if self._format_preprocessor is None:
-            self._format_preprocessor = FormatExcel(path=path)
+        self._format_preprocessor = FormatExcel(path=path)
         return self._format_preprocessor
 
     def normalize(
@@ -137,6 +139,7 @@ class Panel:
         group_columns: Columns = None,
         no_process_columns: Columns = None,
         process_all_characteristics: bool = True,
+        fillna_zero_after_norm: bool = False,
     ) -> Panel:
         """
         Normalizes specified columns of the DataFrame using the chosen method.
@@ -157,6 +160,8 @@ class Panel:
                 normalization. Defaults to None.
             process_all_characteristics (bool, optional): Whether to process all
                 characteristics or not. Defaults to True.
+            fillna_zero_after_norm (bool): If True, fills NaN values with zero after normalization.
+                Defaults to False.
 
         Returns:
             Normalize: The instance of the Normalize class with updated state.
@@ -171,6 +176,7 @@ class Panel:
             group_columns=group_columns,
             no_process_columns=no_process_columns,
             process_all_characteristics=process_all_characteristics,
+            fillna_zero_after_norm=fillna_zero_after_norm,
         ).panel_data
         return self
 
@@ -482,7 +488,7 @@ class Panel:
         groups: Union[int, list[int]],
         sort_type: Literal["independent", "dependent"] = "independent",
         inplace: bool = False,
-    ) -> tuple:
+    ) -> Optional[pd.DataFrame]:
         """Group variables into portfolios based on specified groups.
 
         This method creates portfolios for the specified variables in the panel data.
@@ -526,6 +532,7 @@ class Panel:
         decimal: Optional[int] = None,
         factor_return: bool = False,
         already_grouped: bool = False,
+        is_endog_return: bool = True,
     ) -> tuple:
         """Perform univariate analysis on the specified core variable.
 
@@ -543,6 +550,7 @@ class Panel:
             factor_return (bool): Whether to output factor returns in the analysis. Defaults to False.
             already_grouped (bool): If True, skips the grouping step assuming data has been pre-grouped.
                 Defaults to False.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             tuple: A tuple containing the equal-weighted and value-weighted results DataFrames.
@@ -556,6 +564,7 @@ class Panel:
             decimal=decimal,
             factor_return=factor_return,
             already_grouped=already_grouped,
+            is_endog_return=is_endog_return,
         )
 
     def bivariate_analysis(
@@ -574,6 +583,7 @@ class Panel:
         decimal: Optional[int] = None,
         factor_return: bool = False,
         already_grouped: bool = False,
+        is_endog_return: bool = True,
     ) -> tuple:
         """Perform bivariate analysis on two specified variables.
 
@@ -595,6 +605,7 @@ class Panel:
             factor_return (bool): Whether to output factor returns in the analysis. Defaults to False.
             already_grouped (bool): If True, skips the grouping step assuming data has been pre-grouped.
                 Defaults to False.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             tuple: A tuple containing the equal-weighted and value-weighted results DataFrames.
@@ -612,6 +623,7 @@ class Panel:
             decimal=decimal,
             factor_return=factor_return,
             already_grouped=already_grouped,
+            is_endog_return=is_endog_return,
         )
 
     def fm_reg(
@@ -731,6 +743,10 @@ if __name__ == "__main__":
         # no_process_columns="MktCap",
         # process_all_characteristics=True,
     )
+
+    panel.winsorize(method="winsorize", group_columns="date")
+    pp(panel)
+
     # panel.normalize(
     #     # columns="MktCap",
     #     method="zscore",
@@ -740,9 +756,6 @@ if __name__ == "__main__":
     # )
     # panel.shift(periods=1, drop_original=False)
 
-    panel.winsorize(method="winsorize", group_columns="date")
-    pp(panel)
-
     # summary = panel.summary()
     # pp(summary)
 

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/datasets/dataset.py

@@ -1,7 +1,9 @@
 from importlib import resources
 
-from anomalylab.utils import *
-from anomalylab.utils.imports import *
+import pandas as pd
+from pandas import DataFrame
+
+from anomalylab.utils import pp
 
 
 class DataSet:

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/correlation.py

@@ -1,8 +1,12 @@
-from anomalylab.config import *
+from dataclasses import dataclass
+from typing import Optional
+
+import numpy as np
+from pandas import DataFrame
+
 from anomalylab.empirical.empirical import Empirical
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp, round_to_string
 
 
 @dataclass

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/empirical.py

@@ -1,7 +1,6 @@
-from anomalylab.config import *
+from dataclasses import dataclass
+
 from anomalylab.preprocess.preprocessor import Preprocessor
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
 
 
 @dataclass

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/fm_regression.py

@@ -1,9 +1,27 @@
-from anomalylab.config import *
+import math
+import warnings
+from dataclasses import dataclass
+from functools import partial
+from typing import Literal, Optional
+
+import numpy as np
+import pandas as pd
+import statsmodels.api as sm
+from linearmodels import FamaMacBeth
+from pandas import DataFrame, Series
+
 from anomalylab.empirical.empirical import Empirical
 from anomalylab.preprocess import OutlierHandler
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import (
+    RegModel,
+    RegModels,
+    RegResult,
+    columns_to_list,
+    get_significance_star,
+    pp,
+    round_to_string,
+)
 
 
 @dataclass
@@ -76,12 +94,14 @@ class FamaMacBethRegression(Empirical):
                     raise ValueError(
                         "When calculating the value-weighted industry return, the weight column must be specified!"
                     )
-                func = lambda x: np.average(
-                    x, weights=self.panel_data.df.loc[x.index, weight]
-                )
+
+                def func(x):
+                    return np.average(
+                        x, weights=self.panel_data.df.loc[x.index, weight]
+                    )
             else:
                 raise ValueError(
-                    f"industry_weighed_method must be one of ['value', 'equal']"
+                    "industry_weighed_method must be one of ['value', 'equal']"
                 )
             self.panel_data.df[endog] -= self.panel_data.df.groupby(
                 by=[self.time, industry]

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/persistence.py

@@ -1,11 +1,17 @@
-from importlib import resources
+from importlib import resources  # noqa: F401
+from dataclasses import dataclass
+from typing import Optional
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import seaborn as sns
+from pandas import DataFrame
 
-from anomalylab.config import *
 from anomalylab.empirical.empirical import Empirical
 from anomalylab.preprocess.shift import Shift
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp, round_to_string
 
 
 @dataclass
@@ -72,7 +78,6 @@ class Persistence(Empirical):
         for var in columns:
             all_monthly_corrs = []
             for lag in periods:
-
                 # Store monthly correlations
                 monthly_corrs = []
 

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/portfolio.py

@@ -1,10 +1,16 @@
-from pandas import DataFrame
+import math
+import warnings
+from dataclasses import dataclass
+from typing import Literal, Optional, Union
+
+import numpy as np
+import pandas as pd
+import statsmodels.api as sm
+from pandas import DataFrame, Series
 
-from anomalylab.config import *
 from anomalylab.empirical.empirical import Empirical
 from anomalylab.structure import PanelData, TimeSeries
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import pp, round_to_string
 
 warnings.simplefilter(action="ignore", category=FutureWarning)
 
@@ -116,7 +122,7 @@ class PortfolioAnalysis(Empirical):
         group_col = [self.time]
         for i, var in enumerate(vars):
             if sort_type == "dependent" and i > 0:
-                group_col.append(f"{vars[i-1]}_g{groups[i-1]}")
+                group_col.append(f"{vars[i - 1]}_g{groups[i - 1]}")
                 out_df[f"{var}_g{groups[i]}"] = (
                     out_df.groupby(group_col, observed=False)[var]
                     .apply(
@@ -127,7 +133,7 @@ class PortfolioAnalysis(Empirical):
                         )
                     )
                     .reset_index()
-                    .set_index(f"level_{i+1}")
+                    .set_index(f"level_{i + 1}")
                     .drop(group_col, axis=1)
                 )
             else:
@@ -142,7 +148,7 @@ class PortfolioAnalysis(Empirical):
                         )
                     )
                     .reset_index()
-                    .set_index(f"level_{1}")
+                    .set_index("level_1")
                     .drop(self.time, axis=1)
                 )
 
@@ -157,24 +163,33 @@ class PortfolioAnalysis(Empirical):
         else:
             return out_df
 
-    def _claculate_value(self, df: DataFrame, decimal: Optional[int] = None) -> dict:
+    def _claculate_value(
+        self, df: DataFrame, decimal: Optional[int] = None, is_endog_return: bool = True
+    ) -> dict:
         """Calculate various portfolio performance metrics.
 
         This method computes mean returns, t-values, Sharpe ratios, and model-adjusted alpha and t values.
 
         Args:
             df (DataFrame): The DataFrame containing the relevant data for calculations.
+            decimal (Optional[int]): The number of decimal places for formatting. Defaults to None.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             dict: A dictionary containing computed metrics.
         """
-        stat_dict = self._calculate_mean_and_t_value(df)
-        factors_dict = self._calculate_alpha_and_t_value(df)
-        sharpe_dict = self._calculate_sharpe(df, decimal)
+        stat_dict = self._calculate_mean_and_t_value(df, is_endog_return)
 
-        return {**stat_dict, **factors_dict, **sharpe_dict}
+        if is_endog_return:
+            factors_dict = self._calculate_alpha_and_t_value(df)
+            sharpe_dict = self._calculate_sharpe(df, decimal)
+            return {**stat_dict, **factors_dict, **sharpe_dict}
+
+        return stat_dict
 
-    def _calculate_mean_and_t_value(self, df: DataFrame) -> dict:
+    def _calculate_mean_and_t_value(
+        self, df: DataFrame, is_endog_return: bool = True
+    ) -> dict:
         """Calculate mean and t-value for the dependent variable.
 
         This method computes the mean return and its t-value assuming the null hypothesis
@@ -182,6 +197,7 @@ class PortfolioAnalysis(Empirical):
 
         Args:
             df (DataFrame): The DataFrame containing the relevant data for calculations.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             dict: A dictionary with mean, t-value, and p-value.
@@ -199,7 +215,9 @@ class PortfolioAnalysis(Empirical):
         mean_value = reg.params[0]
         t_value = reg.tvalues[0]
         p_value = reg.pvalues[0]
-        stat_dict["Return"] = mean_value
+
+        key_name = "Return" if is_endog_return else self.endog
+        stat_dict[key_name] = mean_value
         stat_dict["t"] = t_value
         stat_dict["p"] = p_value
 
@@ -284,6 +302,7 @@ class PortfolioAnalysis(Empirical):
         decimal: Optional[int] = None,
         factor_return: bool = False,
         already_grouped: bool = False,
+        is_endog_return: bool = True,
     ) -> tuple:
         """Perform univariate analysis on the specified core variable.
 
@@ -299,6 +318,7 @@ class PortfolioAnalysis(Empirical):
             factor_return (bool): Whether to output factor returns in the analysis. Defaults to False.
             already_grouped (bool): If True, skips the grouping step assuming data has been pre-grouped.
                 Defaults to False.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             tuple: A tuple containing the equal-weighted and value-weighted results DataFrames.
@@ -399,10 +419,20 @@ class PortfolioAnalysis(Empirical):
             results = {}
 
             for key, sr in time_series_dict.items():
-                results[key] = self._claculate_value(sr, decimal=decimal)
+                results[key] = self._claculate_value(
+                    sr, decimal=decimal, is_endog_return=is_endog_return
+                )
 
+            key_name = "Return" if is_endog_return else self.endog
             data = []
             for key, values in results.items():
+                if key_name == core_var:
+                    if key_name in values:
+                        val = values.pop(key_name)
+                        new_values = {f"{key_name}_val": val}
+                        new_values.update(values)
+                        values = new_values
+
                 values[core_var] = key
                 data.append(values)
 
@@ -429,7 +459,7 @@ class PortfolioAnalysis(Empirical):
                 combined_results.iloc[:, i : i + 3] = subset
 
             combined_results = combined_results.loc[
-                :, ~combined_results.columns.str.endswith("p")
+                :, ~combined_results.columns.str.match(r"(^p$|.*-p$)")
             ]
 
             return combined_results
@@ -451,6 +481,7 @@ class PortfolioAnalysis(Empirical):
         decimal: Optional[int] = None,
         factor_return: bool = False,
         already_grouped: bool = False,
+        is_endog_return: bool = True,
     ) -> tuple:
         """Perform bivariate analysis on two specified variables.
 
@@ -470,6 +501,7 @@ class PortfolioAnalysis(Empirical):
             factor_return (bool): Whether to output factor returns in the analysis. Defaults to False.
             already_grouped (bool): If True, skips the grouping step assuming data has been pre-grouped.
                 Defaults to False.
+            is_endog_return (bool): Whether the dependent variable is a return. Defaults to True.
 
         Returns:
             tuple: A tuple containing the equal-weighted and value-weighted results DataFrames.
@@ -544,15 +576,21 @@ class PortfolioAnalysis(Empirical):
 
             return pd.concat([group, sort_diff, sort_avg])
 
+        # Handle potential name collision if endog is same as sort_var or core_var
+        value_col = self.endog
+        if value_col in [sort_var, core_var]:
+            value_col = f"{self.endog}_val"
+
+        ew_ret_d.name = value_col
         ew_ret_d = ew_ret_d.reset_index()
         ew_ret_d = ew_ret_d.pivot(
-            index=[self.time, sort_var], columns=core_var, values=self.endog
+            index=[self.time, sort_var], columns=core_var, values=value_col
         )
 
-        vw_ret_d.name = self.endog
+        vw_ret_d.name = value_col
         vw_ret_d = vw_ret_d.reset_index()
         vw_ret_d = vw_ret_d.pivot(
-            index=[self.time, sort_var], columns=core_var, values=self.endog
+            index=[self.time, sort_var], columns=core_var, values=value_col
         )
 
         ew_ret_d = (
@@ -616,11 +654,21 @@ class PortfolioAnalysis(Empirical):
             results = {}
 
             for key, series in time_series_dict.items():
-                value_dict = self._claculate_value(series, decimal=decimal)
+                value_dict = self._claculate_value(
+                    series, decimal=decimal, is_endog_return=is_endog_return
+                )
                 results[key] = value_dict
 
+            key_name = "Return" if is_endog_return else self.endog
             data = []
             for key, values in results.items():
+                if key_name in [sort_var, core_var]:
+                    if key_name in values:
+                        val = values.pop(key_name)
+                        new_values = {f"{key_name}_val": val}
+                        new_values.update(values)
+                        values = new_values
+
                 values[sort_var] = key[0]
                 values[core_var] = key[1]
                 data.append(values)
@@ -648,7 +696,7 @@ class PortfolioAnalysis(Empirical):
                 combined_results.iloc[:, i : i + 3] = subset
 
             combined_results = combined_results.loc[
-                :, ~combined_results.columns.str.endswith("p")
+                :, ~combined_results.columns.str.match(r"(^p$|.*-p$)")
             ]
 
             def reorder_diff_avg(df: DataFrame) -> DataFrame:
@@ -727,17 +775,22 @@ if __name__ == "__main__":
 
     portfolio = PortfolioAnalysis(
         panel,
-        endog="return",
+        endog="IdioVol",
         weight="MktCap",
-        models=Models,
-        factors_series=time_series,
+        # models=Models,
+        # factors_series=time_series,
     )
 
     # portfolio.GroupN("Illiq", 10, inplace=True)
-    portfolio.GroupN(["MktCap", "Illiq"], [3, 5], sort_type="dependent", inplace=True)
+    portfolio.GroupN(["MktCap", "Illiq"], [5, 5], sort_type="dependent", inplace=True)
 
     uni_ew, uni_vw = portfolio.univariate_analysis(
-        "Illiq", 10, factor_return=False, already_grouped=False
+        "Illiq",
+        5,
+        format=True,
+        # factor_return=False,
+        already_grouped=True,
+        is_endog_return=False,
     )
     pp(uni_ew)
     pp(uni_vw)
@@ -745,13 +798,14 @@ if __name__ == "__main__":
     bi_ew, bi_vw = portfolio.bivariate_analysis(
         "MktCap",
         "Illiq",
-        3,
+        5,
         5,
         False,
-        False,
+        True,
         "dependent",
-        factor_return=False,
+        # factor_return=False,
         already_grouped=True,
+        is_endog_return=False,
     )
     pp(bi_ew)
     pp(bi_vw)

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/empirical/summary.py

@@ -1,12 +1,14 @@
-from anomalylab.config import *
+from dataclasses import dataclass
+from typing import Optional
+
+from pandas import DataFrame, Series
+
 from anomalylab.empirical.empirical import Empirical
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp, round_to_string
 
 
 class Statistics:
-
     @staticmethod
     def mean(series: Series) -> float:
         return series.mean() if not series.isna().all() else None

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/fillna.py

@@ -1,9 +1,14 @@
 from __future__ import annotations
 
+import warnings
+from dataclasses import dataclass
+from typing import Literal, Optional, Union
+
+from pandas import DataFrame, Series
+
 from anomalylab.preprocess.preprocessor import Preprocessor
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, Scalar, columns_to_list, pp
 
 
 @dataclass
@@ -68,7 +73,6 @@ class FillMethod:
 
 @dataclass
 class FillNa(Preprocessor):
-
     def fill(
         self,
         series: Series,
@@ -218,11 +222,11 @@ class FillNa(Preprocessor):
             warnings.warn(message=f"Missing values not found in {fill_columns}.")
         if self.panel_data.normalize:
             warnings.warn(
-                message=f"The data has already been normalized, and missing values have been filled with 0."
+                message="The data has already been normalized, and missing values have been filled with 0."
             )
         if self.panel_data.fillna:
             warnings.warn(
-                message=f"The missing values have already been handled earlier."
+                message="The missing values have already been handled earlier."
             )
 
 

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/normalize.py

@@ -1,9 +1,15 @@
 from __future__ import annotations
 
+import warnings
+from dataclasses import dataclass
+from typing import Literal
+
+import numpy as np
+from pandas import DataFrame, Series
+
 from anomalylab.preprocess.preprocessor import Preprocessor
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp
 
 
 class NormalizeMethod:
@@ -53,7 +59,12 @@ class NormalizeMethod:
         return rescaled_df
 
     @classmethod
-    def call_method(cls, method: str, df: DataFrame) -> DataFrame:
+    def call_method(
+        cls,
+        method: str,
+        df: DataFrame | Series,
+        fillna_zero_after_norm: bool = False,
+    ) -> DataFrame | Series:
         """
         Calls a specified normalization method on the input DataFrame.
 
@@ -66,20 +77,38 @@ class NormalizeMethod:
             cls: The class that is calling this method (NormalizeMethod).
             method (str): The name of the method to call ('zscore' or 'rank').
             df (DataFrame): The input DataFrame to be normalized.
+            fillna_zero_after_norm (bool): If True, fills NaN values with zero after normalization.
+                Defaults to False.
 
         Returns:
-            DataFrame: The normalized DataFrame after applying the specified method.
+            DataFrame: The normalized DataFrame. NaN values are filled with zero
+                if `fillna_zero_after_norm=True` is set.
 
         Raises:
             AttributeError: If the specified method does not exist.
         """
-        if hasattr(cls, method):
-            return getattr(cls, method)(df).fillna(value=0)
-        else:
+        if not hasattr(cls, method):
             raise AttributeError(
                 f"Method '{method}' not found, use 'zscore' or 'rank'."
             )
 
+        normalized_df = getattr(cls, method)(df)
+
+        if fillna_zero_after_norm:
+            normalized_df = normalized_df.fillna(value=0)
+        else:
+            if isinstance(df, Series):
+                if df.isna().all():
+                    warnings.warn(f"Column {df.name} contains only missing values.")
+            else:
+                all_nan_cols = df.columns[df.isna().all()].tolist()
+                if all_nan_cols:
+                    warnings.warn(
+                        f"Columns {all_nan_cols} contain only missing values."
+                    )
+
+        return normalized_df
+
 
 @dataclass
 class Normalize(Preprocessor):
@@ -101,6 +130,7 @@ class Normalize(Preprocessor):
         group_columns: Columns = None,
         no_process_columns: Columns = None,
         process_all_characteristics: bool = True,
+        fillna_zero_after_norm: bool = False,
     ) -> Normalize:
         """
         Normalizes specified columns of the DataFrame using the chosen method.
@@ -121,6 +151,8 @@ class Normalize(Preprocessor):
                 normalization. Defaults to None.
             process_all_characteristics (bool, optional): Whether to process all
                 characteristics or not. Defaults to True.
+            fillna_zero_after_norm (bool): If True, fills NaN values with zero after normalization.
+                Defaults to False.
 
         Returns:
             Normalize: The instance of the Normalize class with updated state.
@@ -144,7 +176,9 @@ class Normalize(Preprocessor):
         # Normalize the selected columns
         self.panel_data.transform(
             columns=columns,
-            func=lambda df: NormalizeMethod.call_method(method=method, df=df),
+            func=lambda df: NormalizeMethod.call_method(
+                method=method, df=df, fillna_zero_after_norm=fillna_zero_after_norm
+            ),
             group_columns=group_columns,
         )
 

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/outliers.py

@@ -1,10 +1,16 @@
 from __future__ import annotations
 
+from dataclasses import dataclass
+from typing import Literal, Optional
+
+import numpy as np
+from pandas import DataFrame, Series
+from scipy.stats.mstats import winsorize as winsorization
+
 from anomalylab.preprocess.preprocessor import Preprocessor
 from anomalylab.preprocess.truncate import truncate as truncation
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp
 
 
 class OutlierMethod:

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/preprocessor.py

@@ -1,8 +1,9 @@
 from __future__ import annotations
 
-from anomalylab.structure import PanelData, TimeSeries
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from abc import ABC
+from dataclasses import dataclass
+
+from anomalylab.structure import PanelData
 
 
 @dataclass

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/shift.py

@@ -1,9 +1,13 @@
 from __future__ import annotations
 
+import warnings
+from dataclasses import dataclass
+
+from pandas import DataFrame
+
 from anomalylab.preprocess.preprocessor import Preprocessor
 from anomalylab.structure import PanelData
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp
 
 
 @dataclass

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/preprocess/truncate.py

@@ -1,11 +1,9 @@
 from __future__ import annotations
 
+import numpy as np
 import numpy.ma as ma
 from scipy._lib._util import _contains_nan
 
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
-
 
 def truncate(
     a,
@@ -113,9 +111,7 @@ def truncate(
         shp = a.shape  # Store the shape of the array
         return _truncate1D(
             a.ravel(), lolim, uplim, loinc, upinc, contains_nan, nan_policy
-        ).reshape(
-            shp
-        )  # Truncate and reshape the array back to its original shape
+        ).reshape(shp)  # Truncate and reshape the array back to its original shape
     else:
         return ma.apply_along_axis(
             _truncate1D, axis, a, lolim, uplim, loinc, upinc, contains_nan, nan_policy

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/structure/data.py

@@ -1,5 +1,10 @@
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+import copy
+from abc import ABC, abstractmethod
+from dataclasses import dataclass
+from typing import Optional
+
+from pandas import DataFrame
+from typing_extensions import Self
 
 
 @dataclass

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/structure/panel_data.py

@@ -1,6 +1,13 @@
+import copy
+import warnings
+from dataclasses import dataclass
+from typing import Callable, Literal, Optional
+
+import pandas as pd
+from pandas import DataFrame
+
 from anomalylab.structure.data import Data
-from anomalylab.utils.imports import *
-from anomalylab.utils.utils import *
+from anomalylab.utils import Columns, columns_to_list, pp
 
 
 @dataclass
@@ -58,8 +65,6 @@ class PanelData(Data):
 
         This method identifies remaining columns as firm characteristics, excluding classifications.
         """
-        if self.is_copy:
-            self.df = copy.deepcopy(self.df)
         self.df[self.id] = self.df[self.id].astype(int)
         if not isinstance(self.df[self.time].dtype, pd.PeriodDtype):
             self.df[self.time] = pd.to_datetime(self.df[self.time], format="ISO8601")
@@ -116,6 +121,9 @@ class PanelData(Data):
             ValueError: If any required columns are missing from the DataFrame.
             ValueError: If there are no firm characteristics remaining after checking.
         """
+        if self.is_copy:
+            self.df = copy.deepcopy(self.df)
+
         # Check for duplicate column names
         duplicated_columns = self.df.columns[self.df.columns.duplicated()].tolist()
         if duplicated_columns:

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/structure/time_series.py

@@ -1,6 +1,12 @@
+import copy
+from dataclasses import dataclass, field
+from typing import Literal
+
+import pandas as pd
+from pandas import DataFrame
+
 from anomalylab.structure.data import Data
-from anomalylab.utils import *
-from anomalylab.utils.imports import *
+from anomalylab.utils import pp
 
 
 @dataclass
@@ -35,8 +41,6 @@ class TimeSeries(Data):
 
         This method renames the time column to a standardized name and identifies remaining columns as factors.
         """
-        if self.is_copy:
-            self.df = copy.deepcopy(self.df)
         if not isinstance(self.df[self.time].dtype, pd.PeriodDtype):
             self.df[self.time] = pd.to_datetime(self.df[self.time], format="ISO8601")
             self.df[self.time] = self.df[self.time].dt.to_period(freq=self.frequency)
@@ -53,6 +57,9 @@ class TimeSeries(Data):
             ValueError: If the time column is missing from the DataFrame.
             ValueError: If there are no additional columns for factor returns.
         """
+        if self.is_copy:
+            self.df = copy.deepcopy(self.df)
+
         # Check for duplicate column names
         duplicated_columns = self.df.columns[self.df.columns.duplicated()].tolist()
         if duplicated_columns:

anomalylab-0.5.0/anomalylab/utils/__init__.py

@@ -0,0 +1,25 @@
+from anomalylab.utils.utils import (
+    Columns,
+    Info,
+    RegModel,
+    RegModels,
+    RegResult,
+    Scalar,
+    columns_to_list,
+    get_significance_star,
+    pp,
+    round_to_string,
+)
+
+__all__: list[str] = [
+    "Scalar",
+    "Columns",
+    "Info",
+    "RegModel",
+    "RegModels",
+    "RegResult",
+    "columns_to_list",
+    "round_to_string",
+    "get_significance_star",
+    "pp",
+]

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/utils/utils.py

@@ -1,4 +1,11 @@
-from anomalylab.utils.imports import *
+from dataclasses import dataclass
+from itertools import chain
+from typing import Any, Optional, TypedDict, Union
+
+from pandas import Series, Timedelta, Timestamp
+from rich import print
+from rich.panel import Panel as rich_Panel
+from rich.pretty import Pretty
 
 Scalar = Union[str, int, float, bool, Timestamp, Timedelta]
 Columns = Optional[list[str] | str]

anomalylab-0.5.0/anomalylab/visualization/__init__.py

@@ -0,0 +1,5 @@
+from anomalylab.visualization.format import FormatExcel
+
+__all__: list[str] = [
+    "FormatExcel",
+]

{anomalylab-0.4.2 → anomalylab-0.5.0}/anomalylab/visualization/format.py

@@ -1,4 +1,9 @@
-from anomalylab.utils.imports import *
+import os
+from dataclasses import dataclass
+from glob import glob
+
+from openpyxl import load_workbook
+from openpyxl.styles import Alignment, Border, Side
 
 
 @dataclass
@@ -45,7 +50,7 @@ class FormatExcel:
         - Creates a thick border for the bottom of the first row.
         """
         thin = Side(border_style="thin", color="000000")
-        thick = Side(border_style="thick", color="000000")
+        thick = Side(border_style="thick", color="000000")  # noqa: F841
 
         for ws in self.wb.worksheets:
             for row in ws.iter_rows():
@@ -104,7 +109,7 @@ class FormatExcel:
                             2 if ord(char) > 127 else 1 for char in str(cell.value)
                         )
                         max_length = max(max_length, cell_length)
-                    except:
+                    except Exception:
                         pass
                 # Adjust for header row
                 # header_cell = ws[f"{col_letter}1"]

{anomalylab-0.4.2 → anomalylab-0.5.0}/setup.py

@@ -2,7 +2,7 @@ from setuptools import find_packages, setup
 
 setup(
     name="AnomalyLab",
-    version="0.4.2",
+    version="0.5.0",
     author="FinPhd",
     author_email="chenhaiwei@stu.sufe.edu.cn",
     description="A Python package for empirical asset pricing analysis.",

anomalylab-0.4.2/anomalylab/config.py

@@ -1 +0,0 @@
-DECIMAL = 2

anomalylab-0.4.2/anomalylab/utils/__init__.py

@@ -1,14 +0,0 @@
-from anomalylab.utils.utils import *
-
-__all__: list[str] = [
-    "Scalar",
-    "Columns",
-    "Info",
-    "RegModel",
-    "RegModels",
-    "RegResult",
-    "columns_to_list",
-    "round_to_string",
-    "get_significance_star",
-    "pp",
-]

anomalylab-0.4.2/anomalylab/utils/imports.py

@@ -1,58 +0,0 @@
-import copy
-import functools
-import math
-import os
-import warnings
-from abc import ABC, ABCMeta, abstractmethod
-from dataclasses import dataclass, field
-from datetime import date, datetime, timedelta, tzinfo
-from functools import partial, wraps
-from glob import glob
-from itertools import chain
-from types import SimpleNamespace
-from typing import (
-    Any,
-    Callable,
-    ClassVar,
-    Generic,
-    Iterable,
-    Literal,
-    Optional,
-    Sequence,
-    TypedDict,
-    TypeVar,
-    Union,
-    get_type_hints,
-)
-
-import matplotlib.pyplot as plt
-import numpy as np
-import pandas as pd
-import seaborn as sns
-import statsmodels.api as sm
-import statsmodels.formula.api as smf
-from deprecated import deprecated
-from linearmodels import FamaMacBeth
-from numpy import float32, float64
-from numpy.typing import NDArray
-from openpyxl import load_workbook
-from openpyxl.styles import Alignment, Border, Side
-from pandas import (
-    DataFrame,
-    DatetimeIndex,
-    Index,
-    Interval,
-    Period,
-    PeriodIndex,
-    Series,
-    Timedelta,
-    Timestamp,
-)
-from pandas.arrays import PeriodArray
-from rich import print
-from rich.panel import Panel as rich_Panel
-from rich.pretty import Pretty, pprint
-from scipy.stats import kurtosis, skew
-from scipy.stats.mstats import winsorize as winsorization
-from tqdm import tqdm
-from typing_extensions import NotRequired, Required, Self

anomalylab-0.4.2/anomalylab/visualization/__init__.py

@@ -1,5 +0,0 @@
-from anomalylab.visualization.format import *
-
-__all__: list[str] = [
-    "FormatExcel",
-]