some new features

.venv/lib/python3.12/site-packages/statsmodels/datasets/committee/__init__.py

@@ -0,0 +1,5 @@
+__all__ = ["load", "load_pandas",
+           "COPYRIGHT", "TITLE", "SOURCE", "DESCRSHORT", "DESCRLONG", "NOTE"]
+from .data import (
+    load, load_pandas,
+    COPYRIGHT, TITLE, SOURCE, DESCRSHORT, DESCRLONG, NOTE)

.venv/lib/python3.12/site-packages/statsmodels/datasets/committee/committee.csv

@@ -0,0 +1,21 @@
+"COMMITTEE","BILLS104","SIZE","SUBS","STAFF","PRESTIGE","BILLS103"
+"Appropriations",6,58,13,109,1,9
+"Budget",23,42,0,39,1,101
+"Rules",44,13,2,25,1,54
+"Ways_and_Means",355,39,5,23,1,542
+"Banking",125,51,5,61,0,101
+"Economic_Educ_Oppor",131,43,5,69,0,158
+"Commerce",271,49,4,79,0,196
+"International_Relations",63,44,3,68,0,40
+"Government_Reform",149,51,7,99,0,72
+"Judiciary",253,35,5,56,0,168
+"Agriculture",81,49,5,46,0,60
+"National_Security",89,55,7,48,0,75
+"Resources",142,44,5,58,0,98
+"TransInfrastructure",155,61,6,74,0,69
+"Science",27,50,4,58,0,25
+"Small_Business",8,43,4,29,0,9
+"Veterans_Affairs",28,33,3,36,0,41
+"House_Oversight",68,12,0,24,0,233
+"Stds_of_Conduct",1,10,0,9,0,0
+"Intelligence",4,16,2,24,0,2

.venv/lib/python3.12/site-packages/statsmodels/datasets/committee/data.py

@@ -0,0 +1,69 @@
+"""First 100 days of the US House of Representatives 1995"""
+from statsmodels.datasets import utils as du
+
+__docformat__ = 'restructuredtext'
+
+COPYRIGHT   = """Used with express permission from the original author,
+who retains all rights."""
+TITLE       = __doc__
+SOURCE      = """
+Jeff Gill's `Generalized Linear Models: A Unifited Approach`
+
+http://jgill.wustl.edu/research/books.html
+"""
+
+DESCRSHORT  = """Number of bill assignments in the 104th House in 1995"""
+
+DESCRLONG   = """The example in Gill, seeks to explain the number of bill
+assignments in the first 100 days of the US' 104th House of Representatives.
+The response variable is the number of bill assignments in the first 100 days
+over 20 Committees.  The explanatory variables in the example are the number of
+assignments in the first 100 days of the 103rd House, the number of members on
+the committee, the number of subcommittees, the log of the number of staff
+assigned to the committee, a dummy variable indicating whether
+the committee is a high prestige committee, and an interaction term between
+the number of subcommittees and the log of the staff size.
+
+The data returned by load are not cleaned to represent the above example.
+"""
+
+NOTE = """::
+
+    Number of Observations - 20
+    Number of Variables - 6
+    Variable name definitions::
+
+        BILLS104 - Number of bill assignments in the first 100 days of the
+                   104th House of Representatives.
+        SIZE     - Number of members on the committee.
+        SUBS     - Number of subcommittees.
+        STAFF    - Number of staff members assigned to the committee.
+        PRESTIGE - PRESTIGE == 1 is a high prestige committee.
+        BILLS103 - Number of bill assignments in the first 100 days of the
+                   103rd House of Representatives.
+
+    Committee names are included as a variable in the data file though not
+    returned by load.
+"""
+
+
+def load_pandas():
+    data = _get_data()
+    return du.process_pandas(data, endog_idx=0)
+
+
+def load():
+    """Load the committee data and returns a data class.
+
+    Returns
+    -------
+    Dataset
+        See DATASET_PROPOSAL.txt for more information.
+    """
+    return load_pandas()
+
+
+def _get_data():
+    data = du.load_csv(__file__, 'committee.csv')
+    data = data.iloc[:, 1:7].astype(float)
+    return data