[FIX] Guard zero case reserve

chainladder/development/outstanding.py

@@ -12,6 +12,7 @@
 
 import numpy as np
 import pandas as pd
+import warnings
 
 if TYPE_CHECKING:
     from chainladder.core import Triangle
@@ -168,17 +169,28 @@ def fit(self, X, y=None, sample_weight=None):
             Development(n_periods=self.case_n_periods).fit(self.X_.sum(0).sum(1)).w_
         )
 
-        case_to_prior_case_naned = self.case_to_prior_case_ * self.case_w_
-        case_to_prior_case_naned.values = num_to_nan(case_to_prior_case_naned.values)
-
-        self.case_ldf_ = self.case_to_prior_case_.mean(2)  # this has the wrong value
-        self.case_ldf_.values = case_to_prior_case_naned.mean(axis=2).values
-
-        paid_to_prior_case_naned = self.paid_to_prior_case_ * self.paid_w_
-        paid_to_prior_case_naned.values = num_to_nan(paid_to_prior_case_naned.values)
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore",
+                category=RuntimeWarning,
+                message=".*empty slice.*",
+            )
+            warnings.filterwarnings(
+                "ignore",
+                category=RuntimeWarning,
+                message=".*All-NaN slice.*",
+            )
+            case_to_prior_case_naned = self.case_to_prior_case_ * self.case_w_
+            case_to_prior_case_naned.values = num_to_nan(
+                case_to_prior_case_naned.values
+            )
+            self.case_ldf_ = case_to_prior_case_naned.mean(axis=2).fillna(1.0)
 
-        self.paid_ldf_ = self.paid_to_prior_case_.mean(2)  # this has the wrong value
-        self.paid_ldf_.values = paid_to_prior_case_naned.mean(axis=2).values
+            paid_to_prior_case_naned = self.paid_to_prior_case_ * self.paid_w_
+            paid_to_prior_case_naned.values = num_to_nan(
+                paid_to_prior_case_naned.values
+            )
+            self.paid_ldf_ = paid_to_prior_case_naned.mean(axis=2).fillzero()
 
         self.ldf_ = self._set_ldf(self.X_).set_backend(backend)
 
@@ -271,9 +283,8 @@ def case_to_prior_case_(self):
             paid_tri = paid_tri.groupby(self.groupby).sum()
             incurred_tri = incurred_tri.groupby(self.groupby).sum()
 
-        out = (incurred_tri - paid_tri).iloc[..., 1:] / (incurred_tri - paid_tri).iloc[
-            ..., :-1
-        ].values
+        prior_case = num_to_nan((incurred_tri - paid_tri).iloc[..., :-1].values)
+        out = (incurred_tri - paid_tri).iloc[..., 1:] / prior_case
         out.is_pattern = True
         out.is_cumulative = False
 
@@ -288,10 +299,8 @@ def paid_to_prior_case_(self):
             paid_tri = paid_tri.groupby(self.groupby).sum()
             incurred_tri = incurred_tri.groupby(self.groupby).sum()
 
-        out = (
-            paid_tri.cum_to_incr().iloc[..., 1:]
-            / (incurred_tri - paid_tri).iloc[..., :-1].values
-        )
+        prior_case = num_to_nan((incurred_tri - paid_tri).iloc[..., :-1].values)
+        out = paid_tri.cum_to_incr().iloc[..., 1:] / prior_case
         out.is_pattern = True
 
         return out

chainladder/development/tests/test_outstanding.py

@@ -345,3 +345,33 @@ def test_unweighted_ratios_preserved_when_n_periods_set():
     assert np.allclose(
         m.case_to_prior_case_.values, m_full.case_to_prior_case_.values, equal_nan=True
     )
+
+
+def test_case_outstanding_zero_case_reserve_warning():
+    import warnings
+
+    xyz = cl.load_sample("friedland_xyz_auto_bi")
+    with warnings.catch_warnings():
+        warnings.filterwarnings("error", category=RuntimeWarning)
+        model = cl.CaseOutstanding(
+            paid_to_incurred=("Paid Claims", "Reported Claims")
+        ).fit(xyz)
+    assert model.paid_ldf_ is not None
+    assert not model.case_ldf_.to_frame().isna().any().any()
+
+
+def test_case_outstanding_synthetic_zero_prior_case_warning():
+    import warnings
+
+    p = cl.load_sample("genins")
+    p.columns = ["paid"]
+    i = p.copy()
+    i.columns = ["incurred"]
+    i.values = p.values + 100
+    # Force case to zero at earlier development maturity
+    i.values[..., 2] = p.values[..., 2]
+    tri = cl.concat((p, i), 1)
+    with warnings.catch_warnings():
+        warnings.filterwarnings("error", category=RuntimeWarning)
+        m_zero = cl.CaseOutstanding(paid_to_incurred=("paid", "incurred")).fit(tri)
+    assert not m_zero.case_ldf_.to_frame().isna().any().any()