tempdisagg 0.2.7__tar.gz → 0.2.9__tar.gz

{tempdisagg-0.2.7/tempdisagg.egg-info → tempdisagg-0.2.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tempdisagg
-Version: 0.2.7
+Version: 0.2.9
 Summary: Temporal disaggregation models in Python
 Author-email: Jaime Vera-Jaramillo <jaimevera1107@gmail.com>
 License-Expression: MIT
@@ -59,7 +59,7 @@ Official indicators often come in low frequency (e.g. yearly GDP), while economi
 | Method(s)                                                               | Description                                                   |
 |-------------------------------------------------------------------------|---------------------------------------------------------------|
 | `ols`                                                                   | Ordinary Least Squares (baseline)                             |
-| `denton`, `denton-opt`                                                  | Denton interpolation with differencing                        |
+| `denton`                                                                | Denton interpolation with differencing                        |
 | `denton-colette`                                                        | Cholette smoother variant from Dagum & Cholette               |
 | `chow-lin`, `chow-lin-opt`, `chow-lin-ecotrim`, `chow-lin-quilis`       | Regression + AR(1) residual modeling                          |
 | `litterman`, `litterman-opt`                                            | Random walk / AR(1) prior models                              |

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/README.md

@@ -36,7 +36,7 @@ Official indicators often come in low frequency (e.g. yearly GDP), while economi
 | Method(s)                                                               | Description                                                   |
 |-------------------------------------------------------------------------|---------------------------------------------------------------|
 | `ols`                                                                   | Ordinary Least Squares (baseline)                             |
-| `denton`, `denton-opt`                                                  | Denton interpolation with differencing                        |
+| `denton`                                                                | Denton interpolation with differencing                        |
 | `denton-colette`                                                        | Cholette smoother variant from Dagum & Cholette               |
 | `chow-lin`, `chow-lin-opt`, `chow-lin-ecotrim`, `chow-lin-quilis`       | Regression + AR(1) residual modeling                          |
 | `litterman`, `litterman-opt`                                            | Random walk / AR(1) prior models                              |

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tempdisagg"
-version = "0.2.7"
+version = "0.2.9"
 description = "Temporal disaggregation models in Python"
 readme = "README.md"
 authors = [

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/tempdisagg/model/tempdisagg_core.py

@@ -130,7 +130,7 @@ class TempDisaggModelCore:
             "litterman-opt": self.models.litterman_opt_estimation,
             "chow-lin-ecotrim": self.models.chow_lin_minrss_ecotrim,
             "chow-lin-quilis": self.models.chow_lin_minrss_quilis,
-            "denton-colette": self.models.denton_cholette_estimation,
+            "denton-cholette": self.models.denton_cholette_estimation,
             "uniform": self.models.uniform_estimation
         }
 

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/tempdisagg/model/tempdisagg_ensemble.py

@@ -99,7 +99,7 @@ class EnsemblePredictor:
             methods = [
                 "ols", "denton", "chow-lin", "litterman", "fernandez", "fast",
                 "chow-lin-opt", "litterman-opt", "chow-lin-ecotrim", "chow-lin-quilis",
-                "denton-opt", "denton-colette", "uniform"
+                "denton-opt", "denton-cholette", "uniform"
             ]
             self.logger.info(f"No methods specified. Using all available: {methods}")
 

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/tempdisagg/model/tempdisagg_fitter.py

@@ -110,7 +110,7 @@ class ModelFitter:
             "litterman-opt": self.models.litterman_opt_estimation,
             "chow-lin-ecotrim": self.models.chow_lin_minrss_ecotrim,
             "chow-lin-quilis": self.models.chow_lin_minrss_quilis,
-            "denton-colette": self.models.denton_cholette_estimation,
+            "denton-cholette": self.models.denton_cholette_estimation,
             "uniform": self.models.uniform_estimation,
         }
 

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/tempdisagg/model/tempdisagg_model.py

@@ -112,7 +112,9 @@ class TempDisaggModel:
             retro_aux_col=retro_aux_col
         ) if self.is_ensemble else None
 
-        self.adjuster = PostEstimation(conversion=conversion)
+        self.adjuster = PostEstimation(conversion=conversion,
+                                       index_col=index_col,
+                                       verbose=verbose)
 
     def fit(self, df, methods=None):
         if self.is_ensemble:

{tempdisagg-0.2.7 → tempdisagg-0.2.9/tempdisagg.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tempdisagg
-Version: 0.2.7
+Version: 0.2.9
 Summary: Temporal disaggregation models in Python
 Author-email: Jaime Vera-Jaramillo <jaimevera1107@gmail.com>
 License-Expression: MIT
@@ -59,7 +59,7 @@ Official indicators often come in low frequency (e.g. yearly GDP), while economi
 | Method(s)                                                               | Description                                                   |
 |-------------------------------------------------------------------------|---------------------------------------------------------------|
 | `ols`                                                                   | Ordinary Least Squares (baseline)                             |
-| `denton`, `denton-opt`                                                  | Denton interpolation with differencing                        |
+| `denton`                                                                | Denton interpolation with differencing                        |
 | `denton-colette`                                                        | Cholette smoother variant from Dagum & Cholette               |
 | `chow-lin`, `chow-lin-opt`, `chow-lin-ecotrim`, `chow-lin-quilis`       | Regression + AR(1) residual modeling                          |
 | `litterman`, `litterman-opt`                                            | Random walk / AR(1) prior models                              |

{tempdisagg-0.2.7 → tempdisagg-0.2.9}/tests/test_prediction.py

@@ -4,20 +4,32 @@ import statsmodels.api as sm
 import pytest
 from tempdisagg import TempDisaggModel
 
+# ---- Fixtures ----
+
 @pytest.fixture(scope="module")
 def disagg_data():
     macro = sm.datasets.macrodata.load_pandas().data
     macro["Index"] = macro["year"].astype(int)
     macro["Grain"] = macro["quarter"].astype(int)
     macro["X"] = macro["realcons"]
-
     col_name = "realgdp"
     gdp_annual = macro.groupby("Index")[col_name].mean().reset_index()
     gdp_annual.columns = ["Index", "y"]
-
     df = macro.merge(gdp_annual, on="Index", how="left")[["Index", "Grain", "y", "X"]].copy()
     return df
 
+@pytest.fixture(scope="module")
+def disagg_data_othername():
+    macro = sm.datasets.macrodata.load_pandas().data
+    macro["index"] = macro["year"].astype(int)
+    macro["grain"] = macro["quarter"].astype(int)
+    macro["x"] = macro["realcons"]
+    col_name = "realgdp"
+    gdp_annual = macro.groupby("index")[col_name].mean().reset_index()
+    gdp_annual.columns = ["index", "y"]
+    df = macro.merge(gdp_annual, on="index", how="left")[["index", "grain", "y", "x"]].copy()
+    return df
+
 # ---- Individual Model Tests ----
 
 def test_individual_predict_truncated(disagg_data):
@@ -29,19 +41,19 @@ def test_individual_predict_truncated(disagg_data):
     assert y_hat.shape[1] == 1
 
 def test_individual_predict_padded(disagg_data):
-    model = TempDisaggModel(method="chow-lin-opt", conversion="average")
+    model = TempDisaggModel(method="denton-cholette", conversion="average")
     model.fit(disagg_data)
     y_hat = model.predict(full=True)
     assert y_hat.shape == (model._df.shape[0], 1)
 
 def test_individual_adjust_padded(disagg_data):
-    model = TempDisaggModel(method="chow-lin-opt", conversion="average")
+    model = TempDisaggModel(method="chow-lin", conversion="average")
     model.fit(disagg_data)
     adjusted = model.adjust_output(full=True)
     assert adjusted.shape == (model._df.shape[0], 1)
 
 def test_individual_adjust_truncated(disagg_data):
-    model = TempDisaggModel(method="chow-lin-opt", conversion="average")
+    model = TempDisaggModel(method="litterman", conversion="average")
     model.fit(disagg_data)
     adjusted = model.adjust_output(full=False)
     assert adjusted.shape[0] < model._df.shape[0]
@@ -77,9 +89,28 @@ def test_ensemble_adjust_truncated(disagg_data):
 
 def test_ensemble_adjust_truncated_no_methods(disagg_data):
     model = TempDisaggModel(method="ensemble", conversion="average")
-    model.fit(disagg_data)
+    model.fit(disagg_data)  # Usa todos los métodos por defecto
     adjusted = model.adjust_output(full=False)
     model.summary()
     model.plot()
     assert adjusted.shape[0] < model._df.shape[0]
     assert adjusted.shape[1] == 1
+
+# ---- Alternative  names ----
+
+def test_individual_othername_predict_truncated(disagg_data_othername):
+    model = TempDisaggModel(method="chow-lin-opt", conversion="average",
+                            index_col="index", grain_col="grain", y_col="y", X_col="x")
+    model.fit(disagg_data_othername)
+    y_hat = model.predict(full=True)
+    assert model._df.shape[0] > disagg_data_othername.shape[0]
+    assert y_hat.shape[0] == model._df.shape[0]
+    assert y_hat.shape[1] == 1
+
+def test_individual_othername_predict_padded(disagg_data_othername):
+    model = TempDisaggModel(method="chow-lin-opt", conversion="average",
+                            index_col="index", grain_col="grain", y_col="y", X_col="x")
+    model.fit(disagg_data_othername)
+    y_hat = model.predict()
+    assert y_hat.shape == (model._df.shape[0], 1)
+