add checks, add docs to crand-using classes

esda/crand.py

@@ -89,8 +89,15 @@ def crand(
         Spatial weights object
     observed : ndarray
         (N,) array with observed values
-    permutations : int
-        Number of permutations for conditional randomisation
+    permutations : int, np.ndarray
+        Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+        conditional random null hypothesis for each site. Permutations
+        might be specified as an array of indices if the user needs to add
+        structure to this conditional permutation null hypothesis. Common
+        reasons to do this include exchangeability violations, which might
+        then require us to shuffle observations within (but not between)
+        groups, or linearity constraints, which may require certain
+        sequences of observation relationships to be preserved.
     keep : Boolean
         If True, store simulation; else do not return randomised statistics
     n_jobs : int

esda/geary_local.py

@@ -42,10 +42,15 @@ def __init__(
                            (default=0.05)
                            Default significance threshold used for
                            creation of labels groups.
-        permutations     : int
-                           (default=999)
-                           number of random permutations for calculation
-                           of pseudo p_values
+        permutations : int, np.ndarray
+            Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+            conditional random null hypothesis for each site. Permutations
+            might be specified as an array of indices if the user needs to add
+            structure to this conditional permutation null hypothesis. Common
+            reasons to do this include exchangeability violations, which might
+            then require us to shuffle observations within (but not between)
+            groups, or linearity constraints, which may require certain
+            sequences of observation relationships to be preserved.
         n_jobs           : int
                            (default=1)
                            Number of cores to be used in the conditional

esda/getisord.py

@@ -257,9 +257,15 @@ class G_Local:  # noqa: N801
         spatial weights instance as W or Graph aligned with y
     transform : {'R', 'B'}
         the type of w, either 'B' (binary) or 'R' (row-standardized)
-    permutations : int
-        the number of random permutations for calculating
-        pseudo p values
+    permutations : int, np.ndarray
+        Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+        conditional random null hypothesis for each site. Permutations
+        might be specified as an array of indices if the user needs to add
+        structure to this conditional permutation null hypothesis. Common
+        reasons to do this include exchangeability violations, which might
+        then require us to shuffle observations within (but not between)
+        groups, or linearity constraints, which may require certain
+        sequences of observation relationships to be preserved.
     star : boolean or float
         whether or not to include focal observation in sums (default: False)
         if the row-transformed weight is provided, then this is the default

esda/join_counts_local.py

@@ -30,9 +30,15 @@ def __init__(
         ----------
         connectivity : W | Graph
             spatial weights instance as W or Graph aligned with y
-        permutations : int
-            number of random permutations for calculation of pseudo
-            p_values
+        permutations : int, np.ndarray
+            Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+            conditional random null hypothesis for each site. Permutations
+            might be specified as an array of indices if the user needs to add
+            structure to this conditional permutation null hypothesis. Common
+            reasons to do this include exchangeability violations, which might
+            then require us to shuffle observations within (but not between)
+            groups, or linearity constraints, which may require certain
+            sequences of observation relationships to be preserved.
         n_jobs : int
             Number of cores to be used in the conditional randomisation. If -1,
             all available cores are used.

esda/join_counts_local_bv.py

@@ -30,8 +30,15 @@ def __init__(
         ----------
         connectivity : W | Graph
             spatial weights instance as W or Graph aligned with y
-        permutations : int
-            number of random permutations for calculation of pseudo p_values
+        permutations : int, np.ndarray
+            Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+            conditional random null hypothesis for each site. Permutations
+            might be specified as an array of indices if the user needs to add
+            structure to this conditional permutation null hypothesis. Common
+            reasons to do this include exchangeability violations, which might
+            then require us to shuffle observations within (but not between)
+            groups, or linearity constraints, which may require certain
+            sequences of observation relationships to be preserved.
         n_jobs : int
             Number of cores to be used in the conditional randomisation.
             If -1, all available cores are used.

esda/join_counts_local_mv.py

@@ -30,8 +30,15 @@ def __init__(
         ----------
         connectivity : W | Graph
             spatial weights instance as W or Graph aligned with y
-        permutations : int
-            number of random permutations for calculation of pseudo p_values
+        permutations : int, np.ndarray
+            Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+            conditional random null hypothesis for each site. Permutations
+            might be specified as an array of indices if the user needs to add
+            structure to this conditional permutation null hypothesis. Common
+            reasons to do this include exchangeability violations, which might
+            then require us to shuffle observations within (but not between)
+            groups, or linearity constraints, which may require certain
+            sequences of observation relationships to be preserved.
         n_jobs : int
             Number of cores to be used in the conditional randomisation. If -1,
             all available cores are used.

esda/moran.py

@@ -891,9 +891,15 @@ class Moran_Local:  # noqa: N801
          "D": doubly-standardized,
          "U": untransformed (general weights),
          "V": variance-stabilizing.
-    permutations : int
-        number of random permutations for calculation of pseudo
-        p_values
+    permutations : int, np.ndarray
+        Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+        conditional random null hypothesis for each site. Permutations
+        might be specified as an array of indices if the user needs to add
+        structure to this conditional permutation null hypothesis. Common
+        reasons to do this include exchangeability violations, which might
+        then require us to shuffle observations within (but not between)
+        groups, or linearity constraints, which may require certain
+        sequences of observation relationships to be preserved.
     geoda_quads : boolean
         (default=False)
         If True use GeoDa scheme: HH=1, LL=2, LH=3, HL=4
@@ -1265,9 +1271,15 @@ class Moran_Local_BV:  # noqa: N801
         "D": doubly-standardized,
         "U": untransformed (general weights),
         "V": variance-stabilizing.
-    permutations   : int
-        number of random permutations for calculation of pseudo
-        p_values
+    permutations : int, np.ndarray
+        Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+        conditional random null hypothesis for each site. Permutations
+        might be specified as an array of indices if the user needs to add
+        structure to this conditional permutation null hypothesis. Common
+        reasons to do this include exchangeability violations, which might
+        then require us to shuffle observations within (but not between)
+        groups, or linearity constraints, which may require certain
+        sequences of observation relationships to be preserved.
     geoda_quads    : boolean
         (default=False)
         If True use GeoDa scheme: HH=1, LL=2, LH=3, HL=4
@@ -1536,9 +1548,15 @@ class Moran_Local_Rate(Moran_Local):  # noqa: N801
         "D": doubly-standardized,
         "U": untransformed (general weights),
         "V": variance-stabilizing.
-    permutations : int
-        number of random permutations for calculation of pseudo
-        p_values
+    permutations : int, np.ndarray
+        Number of permutations for conditional randomisation, or the permutation array itself. Providing an integer will test the
+        conditional random null hypothesis for each site. Permutations
+        might be specified as an array of indices if the user needs to add
+        structure to this conditional permutation null hypothesis. Common
+        reasons to do this include exchangeability violations, which might
+        then require us to shuffle observations within (but not between)
+        groups, or linearity constraints, which may require certain
+        sequences of observation relationships to be preserved.
     geoda_quads : boolean
          (default=False)
          If True use GeoDa scheme: HH=1, LL=2, LH=3, HL=4