skfolio 0.10.0__py3-none-any.whl → 0.10.1__py3-none-any.whl

skfolio/measures/_measures.py

@@ -136,7 +136,9 @@ def variance(
     if sample_weight is None:
         return np.var(returns, ddof=0 if biased else 1, axis=0)
 
-    biased_var = sample_weight @ (returns - mean(returns)) ** 2
+    biased_var = (
+        sample_weight @ (returns - mean(returns, sample_weight=sample_weight)) ** 2
+    )
     if biased:
         return biased_var
     n_eff = 1 / np.sum(sample_weight**2)
@@ -177,7 +179,7 @@ def semi_variance(
         If `returns` is a 2D-array, the result is a ndarray of shape (n_assets,).
     """
     if min_acceptable_return is None:
-        min_acceptable_return = mean(returns)
+        min_acceptable_return = mean(returns, sample_weight=sample_weight)
 
     biased_semi_var = mean(
         np.maximum(0, min_acceptable_return - returns) ** 2, sample_weight=sample_weight

skfolio/prior/_entropy_pooling.py

@@ -589,10 +589,10 @@ class EntropyPooling(BasePrior):
 
         Parameters
         ----------
-        a : ndarray of shape (n_observations, n_constrains)
+        a : ndarray of shape (n_observations, n_constraints)
             Left matrix in `x @ a == b` or `x @ a <= b`.
 
-        a : ndarray of shape (n_observations, n_constrains)
+        a : ndarray of shape (n_observations, n_constraints)
             Right vector in `x @ a == b` or `x @ a <= b`.
 
         Returns
@@ -1029,16 +1029,16 @@ class EntropyPooling(BasePrior):
 
         """
         n_observations, _ = self._returns.shape
-        # Init constrains with sum(p)==1, rescaled by its norm
+        # Init constraints with sum(p)==1, rescaled by its norm
         # Has better convergence than the normalized form done inside the dual.
         a = [np.ones(n_observations).reshape(-1, 1) / np.sqrt(n_observations)]
         b = [np.array([1.0]) / np.sqrt(n_observations)]
         bounds = [(None, None)]
-        for name, constrains in self._constraints.items():
-            if constrains is not None:
-                a.append(constrains[0])
-                b.append(constrains[1])
-                s = constrains[1].size
+        for name, constraints in self._constraints.items():
+            if constraints is not None:
+                a.append(constraints[0])
+                b.append(constraints[1])
+                s = constraints[1].size
                 match name:
                     case "equality" | "cvar_equality":
                         bounds += [(None, None)] * s
@@ -1145,7 +1145,7 @@ class EntropyPooling(BasePrior):
 
         if self._constraints["fixed_equality"] is not None:
             a, b = self._constraints["fixed_equality"]
-            # Relaxe the problem with slack variables with a norm1 penalty to avoid
+            # Relax the problem with slack variables with a norm1 penalty to avoid
             # solver infeasibility that may arise from overly tight constraints from
             # fixing the moments.
             slack = cp.Variable(b.size)

{skfolio-0.10.0.dist-info → skfolio-0.10.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: skfolio
-Version: 0.10.0
+Version: 0.10.1
 Summary: Portfolio optimization built on top of scikit-learn
 Author-email: Hugo Delatte <delatte.hugo@gmail.com>
 Maintainer-email: Hugo Delatte <delatte.hugo@gmail.com>, Matteo Manzi <matteomanzi09@gmail.com>
@@ -385,6 +385,7 @@ Imports
     )
     from skfolio.optimization import (
         MeanRisk,
+        HierarchicalRiskParity,
         NestedClustersOptimization,
         ObjectiveFunction,
         RiskBudgeting,
@@ -584,11 +585,13 @@ Factor Model
 
     factor_prices = load_factors_dataset()
 
-    X, y = prices_to_returns(prices, factor_prices)
-    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, shuffle=False)
+    X, factors = prices_to_returns(prices, factor_prices)
+    X_train, X_test, factors_train, factors_test = train_test_split(
+        X, factors, test_size=0.33, shuffle=False
+    )
 
     model = MeanRisk(prior_estimator=FactorModel())
-    model.fit(X_train, y_train)
+    model.fit(X_train, factors_train)
 
     print(model.weights_)
 
@@ -597,7 +600,6 @@ Factor Model
     print(portfolio.calmar_ratio)
     print(portfolio.summary())
 
-
 Factor Model & Covariance Detoning
 ----------------------------------
 .. code-block:: python
@@ -658,7 +660,7 @@ Combinatorial Purged Cross-Validation
 
     cv = CombinatorialPurgedCV(n_folds=10, n_test_folds=2)
 
-    print(cv.get_summary(X_train))
+    print(cv.summary(X_train))
 
     population = cross_val_predict(model, X_train, cv=cv)
 
@@ -674,7 +676,7 @@ Minimum CVaR Optimization on Synthetic Returns
 .. code-block:: python
 
     vine = VineCopula(log_transform=True, n_jobs=-1)
-    prior = =SyntheticData(distribution_estimator=vine, n_samples=2000)
+    prior = SyntheticData(distribution_estimator=vine, n_samples=2000)
     model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=prior)
     model.fit(X)
     print(model.weights_)
@@ -684,7 +686,7 @@ Stress Test
 -----------
 .. code-block:: python
 
-    vine = VineCopula(log_transform=True, central_assets=["BAC"]  n_jobs=-1)
+    vine = VineCopula(log_transform=True, central_assets=["BAC"], n_jobs=-1)
     vine.fit(X)
     X_stressed = vine.sample(n_samples=10_000, conditioning = {"BAC": -0.2})
     ptf_stressed = model.predict(X_stressed)
@@ -702,7 +704,7 @@ Minimum CVaR Optimization on Synthetic Factors
     )
     factor_model = FactorModel(factor_prior_estimator=factor_prior)
     model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=factor_model)
-    model.fit(X, y)
+    model.fit(X, factors)
     print(model.weights_)
 
 
@@ -713,7 +715,7 @@ Factor Stress Test
     factor_model.set_params(factor_prior_estimator__sample_args=dict(
         conditioning={"QUAL": -0.5}
     ))
-    factor_model.fit(X,y)
+    factor_model.fit(X, factors)
     stressed_dist = factor_model.return_distribution_
     stressed_ptf = model.predict(stressed_dist)
 

{skfolio-0.10.0.dist-info → skfolio-0.10.1.dist-info}/RECORD

@@ -37,7 +37,7 @@ skfolio/distribution/univariate/_selection.py,sha256=6KL4gngiLKwaBUpCDX19ABOkMBz
 skfolio/distribution/univariate/_student_t.py,sha256=GcI4fKp6q5XegfvT_i3AqfWlUMxCq7A5sX6Xsf4pye8,4553
 skfolio/measures/__init__.py,sha256=saJ3tGQyZ5MRtwnj0tVS9L4yvnLJVSdyXVyH5_FyoNo,1683
 skfolio/measures/_enums.py,sha256=S6WOT8NHzm-eMHELuOjngIBupCctCdiTA2BaJlWl-4E,8956
-skfolio/measures/_measures.py,sha256=ysON3rPfcLPSAPY3T1MfkuLPoKU__KN5sRyCzOXe6to,28716
+skfolio/measures/_measures.py,sha256=-pF9DtJqsFliSiv9Ek_7IzytTmoKQFdqCxUl5fdSlYQ,28790
 skfolio/metrics/__init__.py,sha256=ebu5h7Q9X0f3ZZ1VFmAEBPic2sirboKG_zNBHO5abjo,98
 skfolio/metrics/_scorer.py,sha256=L-qct4cby15a4xC4arSaG5__1mxBCQYeMjlrHBIVnSY,4325
 skfolio/model_selection/__init__.py,sha256=BT8VCXW7C4bXI2Oam4amTHOcJVlKxLpkcsHjB63pZHQ,524
@@ -100,7 +100,7 @@ skfolio/prior/__init__.py,sha256=4bi4u7Y-D9vLKb0nxlAXYEZUuYkjPXQcC7qlYUu_DMA,720
 skfolio/prior/_base.py,sha256=aSqyhBYc35RpWq4XpM3UsOu88Bvxbqn7QhctK9bP0I0,2217
 skfolio/prior/_black_litterman.py,sha256=oRH6wUjsL5bkjiNbVtxaIPMNi34rqPp7WBmDbJiklKM,10675
 skfolio/prior/_empirical.py,sha256=zRceQNsbeWdkZHIaFvO91AhZTqkPd8YE2f60cK39M-U,7486
-skfolio/prior/_entropy_pooling.py,sha256=TBczvvauhaV0Uq5NOZXHWc3Ojru-PFNosvbm79RuT9E,59278
+skfolio/prior/_entropy_pooling.py,sha256=x4qd-bU52ZGePwduPNY5fBhWwrNHfI77oWp5x0KyLPc,59285
 skfolio/prior/_factor_model.py,sha256=lbXFsuidDJvLBX7fwp6fXXvgdL3MzkL5lJCx7HEABcA,12241
 skfolio/prior/_opinion_pooling.py,sha256=dBZ8TjlAOKWA9lZFD4DS_PH5HG8iZYgtNrJLPnqwX0o,19055
 skfolio/prior/_synthetic_data.py,sha256=HsxcIQa9lwcYh81Y-o00xsSr9-djgrTQtRTE4SCNpKo,8735
@@ -116,8 +116,8 @@ skfolio/utils/figure.py,sha256=2U0PuHRuza_1N6o_fWD8amNDc0IhgfzM5owFl3zBzwg,5744
 skfolio/utils/sorting.py,sha256=F7gfIBfnulfDUiqvzrlR-pba4PPLJT6NH7-5s4sdRhw,3521
 skfolio/utils/stats.py,sha256=glVHo7rjwy06dl5kkULLOADMrEkVJcfXXAz-1qmYQL4,17005
 skfolio/utils/tools.py,sha256=XQ-bkbhIqBTuv_ZLK-vDnDx8NGFCFvmWoJF8Ui1tj38,23020
-skfolio-0.10.0.dist-info/licenses/LICENSE,sha256=F6Gi-ZJX5BlVzYK8R9NcvAkAsKa7KO29xB1OScbrH6Q,1526
-skfolio-0.10.0.dist-info/METADATA,sha256=bZvW8kQjrwBd2XWFdvZElhNYQaJVzCxkjo0f8Ju7pUU,25053
-skfolio-0.10.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-skfolio-0.10.0.dist-info/top_level.txt,sha256=NXEaoS9Ms7t32gxkb867nV0OKlU0KmssL7IJBVo0fJs,8
-skfolio-0.10.0.dist-info/RECORD,,
+skfolio-0.10.1.dist-info/licenses/LICENSE,sha256=F6Gi-ZJX5BlVzYK8R9NcvAkAsKa7KO29xB1OScbrH6Q,1526
+skfolio-0.10.1.dist-info/METADATA,sha256=LwEyMOUgw9X1asJvXVAVvkNSYeKXvqo35bZsZBFQ28o,25136
+skfolio-0.10.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+skfolio-0.10.1.dist-info/top_level.txt,sha256=NXEaoS9Ms7t32gxkb867nV0OKlU0KmssL7IJBVo0fJs,8
+skfolio-0.10.1.dist-info/RECORD,,