Skip to content

Commit 23fa23c

Browse files
SvenKlaassenSvenKlaassen
committed
update cs example and add to gallery
1 parent bb363c7 commit 23fa23c

File tree

2 files changed

+33
-30
lines changed

2 files changed

+33
-30
lines changed

‎doc/examples/did/py_rep_cs.ipynb

Lines changed: 32 additions & 30 deletions
Original file line numberDiff line numberDiff line change
@@ -4,7 +4,7 @@
44
"cell_type": "markdown",
55
"metadata": {},
66
"source": [
7-
"# Python: Pepeated Cross-Sectional Data with Multiple Time Periods\n",
7+
"# Python: Repeated Cross-Sectional Data with Multiple Time Periods\n",
88
"\n",
99
"In this example, a detailed guide on Difference-in-Differences with multiple time periods using the [DoubleML-package](https://docs.doubleml.org/stable/index.html). The implementation is based on [Callaway and Sant'Anna(2021)](https://doi.org/10.1016/j.jeconom.2020.12.001).\n",
1010
"\n",
@@ -37,9 +37,11 @@
3737
"source": [
3838
"## Data\n",
3939
"\n",
40-
"We will rely on the `make_did_CS2021` DGP, which is inspired by [Callaway and Sant'Anna(2021)](https://doi.org/10.1016/j.jeconom.2020.12.001) (Appendix SC) and [Sant'Anna and Zhao (2020)](https://doi.org/10.1016/j.jeconom.2020.06.003).\n",
40+
"We will rely on the `make_did_cs_CS2021` DGP, which is inspired by [Callaway and Sant'Anna(2021)](https://doi.org/10.1016/j.jeconom.2020.12.001) (Appendix SC) and [Sant'Anna and Zhao (2020)](https://doi.org/10.1016/j.jeconom.2020.06.003).\n",
4141
"\n",
42-
"We will observe `n_obs` units over `n_periods`. Remark that the dataframe includes observations of the potential outcomes `y0` and `y1`, such that we can use oracle estimates as comparisons. "
42+
"We will observe approximately `n_obs` units over `n_periods`. The parameter `lambda_t` determines the probability of observing a unit ``i`` in time period ``t``. The parameter `lambda_t` is set to 0.5 for all time periods, which means that each unit has a 50% chance of being observed in each time period.\n",
43+
"\n",
44+
"Remark that the dataframe includes observations of the potential outcomes `y0` and `y1`, such that we can use oracle estimates as comparisons."
4345
]
4446
},
4547
{
@@ -389,7 +391,7 @@
389391
"The choice `gt_combinations=\"standard\"`, used estimates all possible combinations of $ATT(g,t_\\text{eval})$ via $\\widehat{ATT}(\\mathrm{g},t_\\text{pre},t_\\text{eval})$,\n",
390392
"where the standard choice is $t_\\text{pre} = \\min(\\mathrm{g}, t_\\text{eval}) - 1$ (without anticipation).\n",
391393
"\n",
392-
"Remark that this includes pre-tests effects if $\\mathrm{g} > t_{eval}$, e.g. $\\widehat{ATT}(g=\\text{2025-04}, t_{\\text{pre}}=\\text{2025-01}, t_{\\text{eval}}=\\text{2025-02})$ which estimates the pre-trend from January to February even if the actual treatment occured in April."
394+
"Remark that this includes pre-tests effects if $\\mathrm{g} > t_{eval}$, e.g. $\\widehat{ATT}(g=3, t_{\\text{pre}}=0, t_{\\text{eval}}=1)$ which estimates the pre-trend from time period $0$ to $1$ even if the actual treatment occured in time period $3$."
393395
]
394396
},
395397
{
@@ -630,9 +632,9 @@
630632
"metadata": {},
631633
"outputs": [],
632634
"source": [
633-
"df[\"e\"] = pd.to_datetime(df[\"t\"]).values.astype(\"datetime64[M]\") - \\\n",
634-
" pd.to_datetime(df[\"d\"]).values.astype(\"datetime64[M]\")\n",
635-
"df.groupby(\"e\")[\"ite\"].mean()[1:]"
635+
"df_treated = df[df[\"d\"] != np.inf].copy()\n",
636+
"df_treated[\"e\"] = df_treated[\"t\"] - df_treated[\"d\"]\n",
637+
"df_treated.groupby(\"e\")[\"ite\"].mean().iloc[1:]"
636638
]
637639
},
638640
{
@@ -899,9 +901,11 @@
899901
"cell_type": "markdown",
900902
"metadata": {},
901903
"source": [
902-
"### Selected Combinations\n",
904+
"### Universal Base Period\n",
903905
"\n",
904-
"Instead it is also possible to just submit a list of tuples containing $(\\mathrm{g}, t_\\text{pre}, t_\\text{eval})$ combinations. E.g. only two combinations"
906+
"The option `gt_combinations=\"universal\"` set $t_\\text{pre} = \\mathrm{g} - \\delta - 1$, corresponding to a universal/constant comparison or base period.\n",
907+
"\n",
908+
"Remark that this implies $t_\\text{pre} > t_\\text{eval}$ for all pre-treatment periods (accounting for anticipation). Therefore these effects do not have the same straightforward interpretation as ATT's."
905909
]
906910
},
907911
{
@@ -910,28 +914,19 @@
910914
"metadata": {},
911915
"outputs": [],
912916
"source": [
913-
"gt_dict = {\n",
914-
" \"gt_combinations\": [\n",
915-
" (4.0, 1, 2),\n",
916-
" (4.0, 1, 3),\n",
917-
" ]\n",
918-
"}\n",
919-
"\n",
920-
"dml_obj_all = DoubleMLDIDMulti(dml_data, **(default_args| gt_dict))\n",
921-
"dml_obj_all.fit()\n",
922-
"dml_obj_all.bootstrap(n_rep_boot=5000)\n",
923-
"dml_obj_all.plot_effects()"
917+
"dml_obj_universal = DoubleMLDIDMulti(dml_data, **(default_args| {\"gt_combinations\": \"universal\"}))\n",
918+
"dml_obj_universal.fit()\n",
919+
"dml_obj_universal.bootstrap(n_rep_boot=5000)\n",
920+
"dml_obj_universal.plot_effects()"
924921
]
925922
},
926923
{
927924
"cell_type": "markdown",
928925
"metadata": {},
929926
"source": [
930-
"### Universal Base Period\n",
931-
"\n",
932-
"The option `gt_combinations=\"universal\"` set $t_\\text{pre} = \\mathrm{g} - \\delta - 1$, corresponding to a universal/constant comparison or base period.\n",
927+
"### Selected Combinations\n",
933928
"\n",
934-
"Remark that this implies $t_\\text{pre} > t_\\text{eval}$ for all pre-treatment periods (accounting for anticipation). Therefore these effects do not have the same straightforward interpretation as ATT's."
929+
"Instead it is also possible to just submit a list of tuples containing $(\\mathrm{g}, t_\\text{pre}, t_\\text{eval})$ combinations. E.g. only two combinations"
935930
]
936931
},
937932
{
@@ -940,16 +935,23 @@
940935
"metadata": {},
941936
"outputs": [],
942937
"source": [
943-
"dml_obj_universal = DoubleMLDIDMulti(dml_data, **(default_args| {\"gt_combinations\": \"universal\"}))\n",
944-
"dml_obj_universal.fit()\n",
945-
"dml_obj_universal.bootstrap(n_rep_boot=5000)\n",
946-
"dml_obj_universal.plot_effects()"
938+
"gt_dict = {\n",
939+
" \"gt_combinations\": [\n",
940+
" (4.0, 1, 2),\n",
941+
" (4.0, 1, 3),\n",
942+
" ]\n",
943+
"}\n",
944+
"\n",
945+
"dml_obj_all = DoubleMLDIDMulti(dml_data, **(default_args| gt_dict))\n",
946+
"dml_obj_all.fit()\n",
947+
"dml_obj_all.bootstrap(n_rep_boot=5000)\n",
948+
"dml_obj_all.plot_effects()"
947949
]
948950
}
949951
],
950952
"metadata": {
951953
"kernelspec": {
952-
"display_name": "Python 3",
954+
"display_name": "dml_dev",
953955
"language": "python",
954956
"name": "python3"
955957
},
@@ -963,7 +965,7 @@
963965
"name": "python",
964966
"nbconvert_exporter": "python",
965967
"pygments_lexer": "ipython3",
966-
"version": "3.12.10"
968+
"version": "3.12.8"
967969
}
968970
},
969971
"nbformat": 4,

‎doc/examples/index.rst

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -62,6 +62,7 @@ Difference-in-Differences
6262
did/py_panel_simple.ipynb
6363
did/py_panel.ipynb
6464
did/py_panel_data_example.ipynb
65+
did/py_rep_cs.ipynb
6566

6667

6768
R: Case studies

0 commit comments

Comments
 (0)