mergeron 2025.739439.11__py3-none-any.whl → 2025.739439.13__py3-none-any.whl

mergeron/gen/upp_tests.py

@@ -17,7 +17,7 @@ from .. import (  # noqa
     UPPAggrSelector,
 )
 from ..core import guidelines_boundaries as gbl  # noqa: TID252
-from . import INVResolution, MarketSampleData, UPPTestRegime, UPPTestsCounts
+from . import INVResolution, MarketsData, UPPTestRegime, UPPTestsCounts
 from . import enforcement_stats as esl
 
 __version__ = VERSION
@@ -32,7 +32,7 @@ class INVRESCntsArgs(TypedDict, total=False):
 
 
 def compute_upp_test_counts(
-    _market_data_sample: MarketSampleData,
+    _market_data_sample: MarketsData,
     _upp_test_parms: gbl.HMGThresholds,
     _upp_test_regime: UPPTestRegime,
     /,
@@ -70,7 +70,7 @@ def compute_upp_test_counts(
 
     np.einsum(
         "ij,ij,ij->ij",
-        _market_data_sample.divr_array,
+        _market_data_sample.divratio_array,
         _market_data_sample.pcm_array[:, ::-1],
         _market_data_sample.price_array[:, ::-1] / _market_data_sample.price_array,
         out=guppi_array,
@@ -78,16 +78,18 @@ def compute_upp_test_counts(
 
     np.divide(
         np.einsum(
-            "ij,ij->ij", _market_data_sample.pcm_array, _market_data_sample.divr_array
+            "ij,ij->ij",
+            _market_data_sample.pcm_array,
+            _market_data_sample.divratio_array,
         ),
-        1 - _market_data_sample.divr_array,
+        1 - _market_data_sample.divratio_array,
         out=ipr_array,
     )
 
     np.divide(ipr_array, 1 - _market_data_sample.pcm_array, out=cmcr_array)
 
     (divr_test_vector,) = _compute_test_array_seq(
-        (_market_data_sample.divr_array,),
+        (_market_data_sample.divratio_array,),
         _market_data_sample.frmshr_array,
         _upp_test_regime.divr_aggregator,
     )
@@ -155,78 +157,42 @@ def compute_upp_test_counts(
 
 def _compute_test_array_seq(
     _test_measure_seq: tuple[ArrayDouble, ...],
-    _wt_array: ArrayDouble,
+    _weights: ArrayDouble,
     _aggregator: UPPAggrSelector,
 ) -> tuple[ArrayDouble, ...]:
-    _wt_array = (
-        _wt_array / np.einsum("ij->i", _wt_array)[:, None]
-        if _aggregator
-        in {
-            UPPAggrSelector.CPA,
-            UPPAggrSelector.CPD,
-            UPPAggrSelector.OSA,
-            UPPAggrSelector.OSD,
-        }
-        else np.array([0.5, 0.5], float)
-    )
-
-    match _aggregator:
-        case UPPAggrSelector.AVG:
-            test_array_seq = (
-                1 / 2 * np.einsum("ij->i", _g)[:, None] for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.CPA:
-            test_array_seq = (
-                np.einsum("ij,ij->i", _wt_array[:, ::-1], _g)[:, None]
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.CPD:
-            test_array_seq = (
-                np.sqrt(np.einsum("ij,ij,ij->i", _wt_array[:, ::-1], _g, _g))[:, None]
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.CPG:
-            test_array_seq = (
-                np.expm1(
-                    np.einsum("ij,ij->i", _wt_array[:, ::-1], np.log1p(_g))[:, None]
-                )
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.DIS:
-            test_array_seq = (
-                np.sqrt(1 / 2 * np.einsum("ij,ij->i", _g, _g))[:, None]
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.GMN:
-            test_array_seq = (
-                np.expm1(np.einsum("ij->i", np.log1p(_g))[:, None] / _g.shape[1])
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.MAX:
-            test_array_seq = (_g.max(axis=1, keepdims=True) for _g in _test_measure_seq)
-        case UPPAggrSelector.MIN:
-            test_array_seq = (_g.min(axis=1, keepdims=True) for _g in _test_measure_seq)
-        case UPPAggrSelector.OSA:
-            test_array_seq = (
-                np.einsum("ij,ij->i", _wt_array, _g)[:, None]
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.OSD:
-            test_array_seq = (
-                np.sqrt(np.einsum("ij,ij,ij->i", _wt_array, _g, _g)[:, None])
-                for _g in _test_measure_seq
-            )
-        case UPPAggrSelector.OSG:
-            test_array_seq = (
-                np.expm1(np.einsum("ij,ij->i", _wt_array, np.log1p(_g))[:, None])
-                for _g in _test_measure_seq
-            )
-        case _:
-            raise ValueError("GUPPI/diversion ratio aggregation method is invalid.")
+    if _aggregator in {UPPAggrSelector.CPA, UPPAggrSelector.CPD}:
+        _weights = _weights[:, ::-1] / np.einsum("ij->i", _weights)[:, None]
+    elif _aggregator in {UPPAggrSelector.OSA, UPPAggrSelector.OSD}:
+        _weights = _weights / np.einsum("ij->i", _weights)[:, None]
+    else:
+        _weights = np.array([0.5, 0.5], float)
+
+    # Use weights calculated above to compute average, distance, and geometric mean:
+    if _aggregator in {UPPAggrSelector.AVG, UPPAggrSelector.CPA, UPPAggrSelector.OSA}:
+        test_array_seq = (
+            np.einsum("ij,ij->i", _weights, _g)[:, None] for _g in _test_measure_seq
+        )
+    elif _aggregator in {UPPAggrSelector.DIS, UPPAggrSelector.CPD, UPPAggrSelector.OSD}:
+        test_array_seq = (
+            np.sqrt(np.einsum("ij,ij,ij->i", _weights, _g, _g))[:, None]
+            for _g in _test_measure_seq
+        )
+    elif _aggregator in {UPPAggrSelector.CPG, UPPAggrSelector.GMN, UPPAggrSelector.OSG}:
+        test_array_seq = (
+            np.expm1(np.einsum("ij,ij->i", _weights[:, ::-1], np.log1p(_g)))[:, None]
+            for _g in _test_measure_seq
+        )
+    elif _aggregator == UPPAggrSelector.MAX:
+        test_array_seq = (_g.max(axis=1, keepdims=True) for _g in _test_measure_seq)
+    elif _aggregator == UPPAggrSelector.MIN:
+        test_array_seq = (_g.min(axis=1, keepdims=True) for _g in _test_measure_seq)
+    else:
+        raise ValueError("GUPPI/diversion ratio aggregation method is invalid.")
     return tuple(test_array_seq)
 
 
 if __name__ == "__main__":
     print(
-        "This module defines classes with methods for generating UPP test arrays and UPP test-counts arrays on given data."
+        "This module defines functionsfor generating UPP test arrays "
+        "and UPP test-counts arrays on given data."
     )

mergeron/perks/guidelines_boundary_functions_extra.py

@@ -6,6 +6,7 @@ and may provide improved precision than core functions, but tend to have
 poor performance
 
 """
+
 from __future__ import annotations
 
 from typing import Literal
@@ -15,7 +16,7 @@ from mpmath import mp, mpf  # type: ignore
 from scipy.spatial.distance import minkowski  # type: ignore
 from sympy import lambdify, simplify, solve, symbols  # type: ignore
 
-from .. import DEFAULT_REC_RATIO, VERSION  # noqa: TID252
+from .. import DEFAULT_REC, VERSION  # noqa: TID252
 from ..core import GuidelinesBoundary, MPFloat  # noqa: TID252
 from ..core import guidelines_boundary_functions as gbf  # noqa: TID252
 from . import GuidelinesBoundaryCallable
@@ -98,11 +99,11 @@ def hhi_delta_boundary_qdtr(_dh_val: float = 0.01, /) -> GuidelinesBoundaryCalla
 
 def diversion_share_boundary_qdtr_wtd_avg(
     _delta_star: float = 0.075,
-    _r_val: float = DEFAULT_REC_RATIO,
+    _r_val: float = DEFAULT_REC,
     /,
     *,
     weighting: Literal["own-share", "cross-product-share"] | None = "own-share",
-    recapture_form: Literal["inside-out", "proportional"] = "inside-out",
+    recapture_form: Literal["inside-out", "fixed"] = "inside-out",
 ) -> GuidelinesBoundaryCallable:
     R"""
     Share combinations for the share-weighted average share-ratio boundary.
@@ -112,12 +113,12 @@ def diversion_share_boundary_qdtr_wtd_avg(
     _delta_star
         Diversion share, :math:`\overline{d} / \overline{r}` or :math:`\overline{g} / (m^* \cdot \overline{r})`.
     _r_val
-        Recapture ratio.
+        Recapture rate.
     weighting
         Whether "own-share" or "cross-product-share" (or None for simple, unweighted average)
     recapture_form
-        Whether recapture-ratio is MNL-consistent ("inside-out") or has fixed
-        value for both merging firms ("proportional").
+        Whether recapture rate is share-proportional ("inside-out") or has fixed
+        value for both merging firms ("fixed").
 
     Returns
     -------
@@ -211,12 +212,12 @@ def diversion_share_boundary_qdtr_wtd_avg(
 
 def diversion_share_boundary_distance(
     _delta_star: float = 0.075,
-    _r_val: float = DEFAULT_REC_RATIO,
+    _r_val: float = DEFAULT_REC,
     /,
     *,
     agg_method: Literal["arithmetic mean", "distance"] = "arithmetic mean",
     weighting: Literal["own-share", "cross-product-share"] | None = "own-share",
-    recapture_form: Literal["inside-out", "proportional"] = "inside-out",
+    recapture_form: Literal["inside-out", "fixed"] = "inside-out",
     dps: int = 5,
 ) -> GuidelinesBoundary:
     R"""
@@ -233,14 +234,14 @@ def diversion_share_boundary_distance(
     _delta_star
         Diversion share, :math:`\overline{d} / \overline{r}` or :math:`\overline{g} / (m^* \cdot \overline{r})`.
     _r_val
-        Recapture ratio.
+        Recapture rate.
     agg_method
         Whether "arithmetic mean" or "distance".
     weighting
         Whether "own-share" or "cross-product-share".
     recapture_form
-        Whether recapture-ratio is MNL-consistent ("inside-out") or has fixed
-        value for both merging firms ("proportional").
+        Whether recapture rate is share-proportional ("inside-out") or has fixed
+        value for both merging firms ("fixed").
     dps
         Number of decimal places for rounding returned shares and area.
 
@@ -292,13 +293,9 @@ def diversion_share_boundary_distance(
 
             match agg_method:
                 case "arithmetic mean":
-                    delta_test = minkowski(
-                        (de_1, de_2), (0.0, 0.0), p=1, w=weights_i
-                    )
+                    delta_test = minkowski((de_1, de_2), (0.0, 0.0), p=1, w=weights_i)
                 case "distance":
-                    delta_test = minkowski(
-                        (de_1, de_2), (0.0, 0.0), p=2, w=weights_i
-                    )
+                    delta_test = minkowski((de_1, de_2), (0.0, 0.0), p=2, w=weights_i)
 
             _test_flag, _incr_decr = (
                 (delta_test > _delta_star, -1)
@@ -366,10 +363,10 @@ def diversion_share_boundary_distance(
 
 def diversion_share_boundary_xact_avg_mp(
     _delta_star: float = 0.075,
-    _r_val: float = DEFAULT_REC_RATIO,
+    _r_val: float = DEFAULT_REC,
     /,
     *,
-    recapture_form: Literal["inside-out", "proportional"] = "inside-out",
+    recapture_form: Literal["inside-out", "fixed"] = "inside-out",
     dps: int = 5,
 ) -> GuidelinesBoundary:
     R"""
@@ -400,7 +397,7 @@ def diversion_share_boundary_xact_avg_mp(
             ylabel=s_2
         )
 
-        # recapture_form = "proportional"
+        # recapture_form = "fixed"
         sag = solve((s_2/(1 - s_1)) + (s_1/(1 - s_2)) - 2 * d_hat, s_2)[0]
         symplot(
             sag,
@@ -413,10 +410,10 @@ def diversion_share_boundary_xact_avg_mp(
     _delta_star
         Diversion share, :math:`\overline{d} / \overline{r}` or :math:`\overline{g} / (m^* \cdot \overline{r})`.
     _r_val
-        Recapture ratio.
+        Recapture rate.
     recapture_form
-        Whether recapture-ratio is MNL-consistent ("inside-out") or has fixed
-        value for both merging firms ("proportional").
+        Whether recapture rate is share-proportional ("inside-out") or has fixed
+        value for both merging firms ("fixed").
     dps
         Number of decimal places for rounding returned shares.
 
@@ -520,14 +517,14 @@ def diversion_share_boundary_xact_avg_mp(
 # this function is about half as fast as the manual one! ... and a touch less precise
 def _diversion_share_boundary_wtd_avg_autoroot(
     _delta_star: float = 0.075,
-    _r_val: float = DEFAULT_REC_RATIO,
+    _r_val: float = DEFAULT_REC,
     /,
     *,
     agg_method: Literal[
         "arithmetic mean", "geometric mean", "distance"
     ] = "arithmetic mean",
     weighting: Literal["own-share", "cross-product-share", None] = "own-share",
-    recapture_form: Literal["inside-out", "proportional"] = "inside-out",
+    recapture_form: Literal["inside-out", "fixed"] = "inside-out",
     dps: int = 5,
 ) -> GuidelinesBoundary:
     R"""
@@ -538,14 +535,14 @@ def _diversion_share_boundary_wtd_avg_autoroot(
     _delta_star
         Diversion share, :math:`\overline{d} / \overline{r}` or :math:`\overline{g} / (m^* \cdot \overline{r})`.
     _r_val
-        Recapture ratio.
+        Recapture rate.
     agg_method
         Whether "arithmetic mean", "geometric mean", or "distance".
     weighting
         Whether "own-share" or "cross-product-share"  (or None for simple, unweighted average).
     recapture_form
-        Whether recapture-ratio is MNL-consistent ("inside-out") or has fixed
-        value for both merging firms ("proportional").
+        Whether recapture rate is share-proportional ("inside-out") or has fixed
+        value for both merging firms ("fixed").
     dps
         Number of decimal places for rounding returned shares and area.
 
@@ -588,7 +585,7 @@ def _diversion_share_boundary_wtd_avg_autoroot(
             (s_1, 0.0, d_hat / (1 + d_hat)), ylabel=s_2
         )
 
-        # recapture_form == "proportional"
+        # recapture_form == "fixed"
         oswag = solve(
             s_1 * s_2 / (1 - s_1)
             + s_2 * s_1 / (1 - s_2)

{mergeron-2025.739439.11.dist-info → mergeron-2025.739439.13.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: mergeron
-Version: 2025.739439.11
+Version: 2025.739439.13
 Summary: Python for analyzing merger enforcement policy
 License: MIT
 Keywords: merger enforcement policy,merger guidelines,merger screening,enforcement presumptions,concentration standards,diversion ratio,upward pricing pressure,GUPPI
@@ -26,7 +26,6 @@ Requires-Dist: joblib (>=1.5.1)
 Requires-Dist: lxml (>=6.0.0)
 Requires-Dist: matplotlib (>=3.10.3)
 Requires-Dist: mpmath (>=1.3.0)
-Requires-Dist: numexpr (>=2.11.0,<3.0.0)
 Requires-Dist: python-calamine (>=0.4.0)
 Requires-Dist: ruamel-yaml (>=0.18.14)
 Requires-Dist: scipy (>=1.16.0)

mergeron-2025.739439.13.dist-info/RECORD

@@ -0,0 +1,21 @@
+mergeron/__init__.py,sha256=7EfQoQmwD1jpLMoPnuXOjraGsAmq44OP39XMiqoadxM,6736
+mergeron/core/__init__.py,sha256=zJcc50CAwm0oNKwk4p-P7oRwVfB_Fp4u0Do5dUeUXfI,3027
+mergeron/core/empirical_margin_distribution.py,sha256=61U-KLB563BPWM5zWyWp82c4PhcsAG-IKI0WWYGjBKg,11740
+mergeron/core/ftc_merger_investigations_data.py,sha256=oM4cs2PnyeSwyV1LOE_EYCUEzCKPm7lnCGxLIc6JQY8,28820
+mergeron/core/guidelines_boundaries.py,sha256=Z8rZvhHrxXBgrLGFpb6yldc8h3lN9rGtGj4yu-fyVBA,15450
+mergeron/core/guidelines_boundary_functions.py,sha256=fdWbqOb3Khz0OkbSTO7amE1q_ao3puZY5tEzj0p4h1o,30695
+mergeron/core/pseudorandom_numbers.py,sha256=CFp-8eu0q2g-81LA0k2oCFltmp6Er7EkrAkoG19G7Os,10138
+mergeron/data/__init__.py,sha256=SAFkR23RBM0zwGam2TeWmw08oHAKmU2YF-Nygj73ies,1845
+mergeron/data/damodaran_margin_data_serialized.zip,sha256=Wc1v9buSrYTWWAravG8W9nPbgsU07zMtSAR2RvMQU5s,623482
+mergeron/data/ftc_merger_investigations_data.zip,sha256=tiB2TLFyS9LMSFIv8DBA_oEEx12DU4MyjHni4NlsRMU,24002
+mergeron/gen/__init__.py,sha256=6xUhaG4kWj2Qx8hLjgjupFWcJ0ZzAKDY9TN7mAFrANI,23880
+mergeron/gen/data_generation.py,sha256=cZW3Dc6bNiBUPXjTDHZDwTc6x1sxXq2STCzfsvk6_tw,17638
+mergeron/gen/data_generation_functions.py,sha256=SWzZ3I7ulkGBcL2F5CCKw2IvCm_wEplvqBasnSjSyU0,26129
+mergeron/gen/enforcement_stats.py,sha256=V3ZeVv-iLFUuKPeM503cMKiVVaYeGVrA_6lInAdXA5w,14387
+mergeron/gen/upp_tests.py,sha256=v-tnhQ85j8zL-TTE52GC61GEZSRFfdCkjaitVQIz0FI,6464
+mergeron/perks/__init__.py,sha256=gGRIuRc7I6OuWLzwSiSZSIE0PEoxAy2DRFWg0VVLlbE,484
+mergeron/perks/guidelines_boundary_functions_extra.py,sha256=q-Cqk9t5oj4yiAsmZJcsfrH434oGvza4YVspFYpdV0g,22113
+mergeron/py.typed,sha256=frcCV1k9oG9oKj3dpUqdJg1PxRT2RSN_XKdLCPjaYaY,2
+mergeron-2025.739439.13.dist-info/METADATA,sha256=pW4fkwJwFD_eALAZ-TSIryYrJu1AMfczHcoDq1u9LdE,4167
+mergeron-2025.739439.13.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+mergeron-2025.739439.13.dist-info/RECORD,,

mergeron-2025.739439.11.dist-info/RECORD

@@ -1,21 +0,0 @@
-mergeron/__init__.py,sha256=h6iafamitU2GozjC36LB3hqIfiNPehoVj5ME7qlJIWg,6752
-mergeron/core/__init__.py,sha256=zJcc50CAwm0oNKwk4p-P7oRwVfB_Fp4u0Do5dUeUXfI,3027
-mergeron/core/empirical_margin_distribution.py,sha256=61U-KLB563BPWM5zWyWp82c4PhcsAG-IKI0WWYGjBKg,11740
-mergeron/core/ftc_merger_investigations_data.py,sha256=oM4cs2PnyeSwyV1LOE_EYCUEzCKPm7lnCGxLIc6JQY8,28820
-mergeron/core/guidelines_boundaries.py,sha256=PvSOLB6KlXu9KX6NdpStuRluP6nrZqUzSHT6usa5600,15630
-mergeron/core/guidelines_boundary_functions.py,sha256=mW5jW7qgn1Z4wMY3P_l86jtJoSke_Wcnd-JKxoWmYKI,30838
-mergeron/core/pseudorandom_numbers.py,sha256=-mPveXjJJ446NrBMAmWIa2jI6j0Px0xcCJTGEEsn3bo,10149
-mergeron/data/__init__.py,sha256=SAFkR23RBM0zwGam2TeWmw08oHAKmU2YF-Nygj73ies,1845
-mergeron/data/damodaran_margin_data_serialized.zip,sha256=Wc1v9buSrYTWWAravG8W9nPbgsU07zMtSAR2RvMQU5s,623482
-mergeron/data/ftc_merger_investigations_data.zip,sha256=tiB2TLFyS9LMSFIv8DBA_oEEx12DU4MyjHni4NlsRMU,24002
-mergeron/gen/__init__.py,sha256=apcCVC29yVjIM1yLM-wVpwrIwfi3BvJp_fPThe0Bgoc,23864
-mergeron/gen/data_generation.py,sha256=pdQl_uBENCjVt2aPtAcus2md4Qx-j5rOWGJJAzdnlO0,17617
-mergeron/gen/data_generation_functions.py,sha256=hXUq5D2CIUkM4_NdGXiOb4XATYwIUeUcCGadzQGDqLw,26126
-mergeron/gen/enforcement_stats.py,sha256=etTax-sBSn8DveF-IxuBJDdX0XSBD6oFU9vaZe6cYks,14387
-mergeron/gen/upp_tests.py,sha256=gRJISQ2jGmIDmFOvaTIkvYooI4mK-QbgkfgL46RrRio,7445
-mergeron/perks/__init__.py,sha256=gGRIuRc7I6OuWLzwSiSZSIE0PEoxAy2DRFWg0VVLlbE,484
-mergeron/perks/guidelines_boundary_functions_extra.py,sha256=lbqjmeLjrqXBm7dEmhiMcqzlv89cDlmyrG3TDzUFSEY,22296
-mergeron/py.typed,sha256=frcCV1k9oG9oKj3dpUqdJg1PxRT2RSN_XKdLCPjaYaY,2
-mergeron-2025.739439.11.dist-info/METADATA,sha256=na0otI3x199j_ks6Qr6oigSYe8aZ7NL4FND9L_417zY,4208
-mergeron-2025.739439.11.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-mergeron-2025.739439.11.dist-info/RECORD,,