Address code review feedback for PR #68

- Remove unused imports (hashlib, os, Optional) from datasets.py
- Clear notebook outputs for cleaner version control
- Improve seed comments for better clarity on reproducibility

Author: claude

diff_diff/datasets.py

@@ -8,11 +8,9 @@
 for subsequent use.
 """
 
-import hashlib
-import os
 from io import StringIO
 from pathlib import Path
-from typing import Dict, Optional
+from typing import Dict
 from urllib.error import HTTPError, URLError
 from urllib.request import urlopen
 
@@ -162,7 +160,7 @@ def _construct_card_krueger_data() -> pd.DataFrame:
     Uses aggregated data that preserves the key DiD estimates.
     """
     # Representative sample based on published summary statistics
-    np.random.seed(1994)  # Year of publication
+    np.random.seed(1994)  # Card-Krueger publication year, for reproducibility
 
     stores = []
     store_id = 1
@@ -307,7 +305,7 @@ def _construct_castle_doctrine_data() -> pd.DataFrame:
 
     This is a fallback when the online source is unavailable.
     """
-    np.random.seed(2013)  # Year of Cheng-Hoekstra publication
+    np.random.seed(2013)  # Cheng-Hoekstra publication year, for reproducibility
 
     # States and their Castle Doctrine adoption years
     # 0 = never adopted during the study period
@@ -456,7 +454,7 @@ def _construct_divorce_laws_data() -> pd.DataFrame:
 
     This is a fallback when the online source is unavailable.
     """
-    np.random.seed(2006)  # Year of Stevenson-Wolfers
+    np.random.seed(2006)  # Stevenson-Wolfers publication year, for reproducibility
 
     # State adoption years for unilateral divorce (from Wolfers 2006)
     # 0 = never adopted or adopted before 1968
@@ -603,7 +601,7 @@ def _construct_mpdta_data() -> pd.DataFrame:
 
     This replicates the simulated dataset used in Callaway-Sant'Anna tutorials.
     """
-    np.random.seed(2021)  # Year of CS publication
+    np.random.seed(2021)  # Callaway-Sant'Anna publication year, for reproducibility
 
     n_counties = 500
     years = [2003, 2004, 2005, 2006, 2007]

docs/tutorials/09_real_world_examples.ipynb

@@ -464,7 +464,24 @@
    "id": "cell-22",
    "metadata": {},
    "outputs": [],
-   "source": "# Callaway-Sant'Anna estimation\ncs = CallawaySantAnna(\n    control_group='never_treated',\n    n_bootstrap=199,\n    seed=42\n)\n\nresults_cs = cs.fit(\n    castle,\n    outcome='homicide_rate',\n    unit='state',\n    time='year',\n    first_treat='first_treat'\n)\n\nprint(results_cs.summary())"
+   "source": [
+    "# Callaway-Sant'Anna estimation\n",
+    "cs = CallawaySantAnna(\n",
+    "    control_group='never_treated',\n",
+    "    n_bootstrap=199,\n",
+    "    seed=42\n",
+    ")\n",
+    "\n",
+    "results_cs = cs.fit(\n",
+    "    castle,\n",
+    "    outcome='homicide_rate',\n",
+    "    unit='state',\n",
+    "    time='year',\n",
+    "    first_treat='first_treat'\n",
+    ")\n",
+    "\n",
+    "print(results_cs.summary())"
+   ]
   },
   {
    "cell_type": "code",
@@ -647,7 +664,24 @@
    "id": "cell-32",
    "metadata": {},
    "outputs": [],
-   "source": "# Callaway-Sant'Anna estimation\ncs_divorce = CallawaySantAnna(\n    control_group='never_treated',\n    n_bootstrap=199,\n    seed=42\n)\n\nresults_divorce = cs_divorce.fit(\n    divorce,\n    outcome='divorce_rate',\n    unit='state',\n    time='year',\n    first_treat='first_treat'\n)\n\nprint(results_divorce.summary())"
+   "source": [
+    "# Callaway-Sant'Anna estimation\n",
+    "cs_divorce = CallawaySantAnna(\n",
+    "    control_group='never_treated',\n",
+    "    n_bootstrap=199,\n",
+    "    seed=42\n",
+    ")\n",
+    "\n",
+    "results_divorce = cs_divorce.fit(\n",
+    "    divorce,\n",
+    "    outcome='divorce_rate',\n",
+    "    unit='state',\n",
+    "    time='year',\n",
+    "    first_treat='first_treat'\n",
+    ")\n",
+    "\n",
+    "print(results_divorce.summary())"
+   ]
   },
   {
    "cell_type": "code",
@@ -791,4 +825,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}