Equations to define causal graph including causal mechanism

.github/workflows/advanced-on-demand.yml

@@ -17,7 +17,7 @@ jobs:
     if: ${{ github.event.comment.body == 'test advanced'}}
     steps:
       - name: Message Initiating
-        uses: actions/github-script@v6
+        uses: actions/github-script@v7
         with:
           script: |
             github.issues.createComment({
@@ -28,7 +28,7 @@ jobs:
             });
       - name: Get PR SHA
         id: sha
-        uses: actions/github-script@v6
+        uses: actions/github-script@v7
         with:
           result-encoding: string
           script: |
@@ -69,7 +69,7 @@ jobs:
 
       - name: Message success
         if: ${{ success() }}
-        uses: actions/github-script@v6
+        uses: actions/github-script@v7
         with:
           script: |
             github.issues.createComment({
@@ -80,7 +80,7 @@ jobs:
             });
       - name: Message failure
         if: ${{ failure() }}
-        uses: actions/github-script@v6
+        uses: actions/github-script@v7
         with:
           script: |
             github.issues.createComment({

docs/source/cite.rst

@@ -1,7 +1,7 @@
 Citing this package
 -------------------
 
-If you find DoWhy useful for your work, please cite the following two references:
+If you find DoWhy useful for your work, please cite **both** of the following two references:
 
 - Amit Sharma, Emre Kiciman. DoWhy: An End-to-End Library for Causal Inference. 2020. https://arxiv.org/abs/2011.04216
 - Patrick Blöbaum, Peter Götz, Kailash Budhathoki, Atalanti A. Mastakouri, Dominik Janzing. DoWhy-GCM: An extension of DoWhy for causal inference in graphical causal models. 2022. https://arxiv.org/abs/2206.06821

docs/source/example_notebooks/2022 First Day.csv

@@ -0,0 +1,2 @@
+Date,Shopping Event?,Ad Spend,Page Views,Unit Price,Sold Units,Revenue,Operational Cost,Profit
+2022-01-01,False,1392.3358280937282,10018,899.1,2756,2477919.6,1879401.202414113,598518.397585887

docs/source/example_notebooks/2022 First Quarter.csv

@@ -0,0 +1,91 @@
+Date,Shopping Event?,Ad Spend,Page Views,Unit Price,Sold Units,Revenue,Operational Cost,Profit
+2022-01-01,False,1392.3358280937282,10018,899.1,2756,2477919.6,1879401.202414113,598518.397585887
+2022-01-02,False,1391.3913808337466,10038,899.1,2726,2450946.6,1864399.9104804778,586546.6895195222
+2022-01-03,False,1293.990843338987,9941,899.1,2656,2388009.6,1829299.397053411,558710.2029465891
+2022-01-04,False,1212.5903975212161,9950,899.1,2741,2464433.1,1871716.1029819865,592716.9970180136
+2022-01-05,False,1182.4698371589263,9764,899.1,2697,2424872.7,1849701.0618950732,575171.638104927
+2022-01-06,False,1467.445551067856,10207,899.1,2738,2461735.8000000003,1870468.5168295426,591267.2831704577
+2022-01-07,True,2141.808717727631,17412,831.3573114052718,5643,4691349.308259949,3323643.689932692,1367705.6183272572
+2022-01-08,False,1234.829256168011,10071,899.1,2723,2448249.3000000003,1862749.2170296635,585500.0829703368
+2022-01-09,False,1376.020055104923,10129,899.1,2688,2416780.8000000003,1845378.2985341656,571402.5014658347
+2022-01-10,False,1340.02761092149,9984,899.1,2794,2512085.4,1898350.3293293247,613735.0706706753
+2022-01-11,False,1273.287440057397,9863,899.1,2696,2423973.6,1849280.0494022565,574693.5505977436
+2022-01-12,False,1141.693146346015,9916,899.1,2759,2480616.9,1880646.1331021162,599970.7668978837
+2022-01-13,False,1155.81122050363,10173,899.1,2831,2545352.1,1916660.5228116934,628691.5771883067
+2022-01-14,False,1378.6091490867877,10145,899.1,2825,2539957.5,1913893.7435477874,626063.7564522126
+2022-01-15,False,1293.478645787744,10082,899.1,2770,2490507.0,1886306.3993566553,604200.6006433447
+2022-01-16,False,1235.8202923804415,10099,899.1,2860,2571426.0,1931237.6707989252,640188.3292010748
+2022-01-17,False,1249.9174124330566,9907,899.1,2717,2442854.7,1859750.6417620734,583104.0582379268
+2022-01-18,False,1271.47439232518,9888,899.1,2677,2406890.7,1839773.284036592,567117.4159634081
+2022-01-19,False,1373.2226475584066,9764,899.1,2696,2423973.6,1849376.4607885955,574597.1392114046
+2022-01-20,False,1340.3791158099252,10054,899.1,2768,2488708.8000000003,1885345.0795596098,603363.7204403905
+2022-01-21,False,1128.1931917991249,9924,899.1,2764,2485112.4,1883130.7364539118,601981.6635460882
+2022-01-22,False,1162.9911436309628,9648,899.1,2627,2361935.7,1814674.633357068,547261.0666429321
+2022-01-23,False,1128.2222747689384,9842,899.1,2545,2288209.5,1773630.7895691658,514578.7104308342
+2022-01-24,False,1113.8360503017068,9802,899.1,2649,2381715.9,1825615.6047651214,556100.2952348785
+2022-01-25,False,1410.401967178016,10229,899.1,2770,2490507.0,1886417.1431889622,604089.8568110378
+2022-01-26,False,1440.3835630651943,9978,899.1,2791,2509388.1,1896940.9275888735,612447.1724111266
+2022-01-27,True,2988.4214329976066,17915,831.9998636067619,6240,5191679.148906195,3622996.5890260655,1568682.559880129
+2022-01-28,False,1487.5363138089551,9990,899.1,2741,2464433.1,1871993.5629440963,592439.5370559038
+2022-01-29,False,1445.4390808945325,9964,899.1,2696,2423973.6,1849447.010754491,574526.5892455091
+2022-01-30,False,1444.1540175500318,10197,899.1,2770,2490507.0,1886446.0023766824,604060.9976233176
+2022-01-31,False,1246.8737599731749,10033,899.1,2745,2468029.5,1873754.1199982178,594275.3800017822
+2022-02-01,False,1481.7388652417399,10066,899.1,2730,2454543.0,1866503.6552066077,588039.3447933923
+2022-02-02,False,1399.5719382535456,9997,899.1,2704,2431166.4,1853399.9672306762,577766.4327693237
+2022-02-03,False,1241.4769716430899,10070,899.1,2704,2431166.4,1853247.3622717985,577919.0377282014
+2022-02-04,False,1455.5331575779842,10091,899.1,2760,2481516.0,1881466.3955227518,600049.6044772482
+2022-02-05,False,1101.6449199860515,9895,899.1,2717,2442854.7,1859602.5783715304,583252.1216284698
+2022-02-06,False,1328.6364164935533,9962,899.1,2817,2532764.7,1909837.6106419214,622927.0893580788
+2022-02-07,False,1308.069117200877,10016,899.1,2775,2495002.5,1888820.615166294,606181.8848337061
+2022-02-08,False,1132.7159762121048,9861,899.1,2659,2390706.9,1830632.84014718,560074.0598528199
+2022-02-09,False,1444.9425958010747,9861,899.1,2716,2441955.6,1859473.3602692473,582482.2397307528
+2022-02-10,False,1212.684467045,9972,899.1,2750,2472525.0,1876230.156967209,596294.843032791
+2022-02-11,False,1250.5400317812196,9750,899.1,2718,2443753.8000000003,1860258.2671390085,583495.5328609918
+2022-02-12,False,1290.0613657717636,9946,899.1,2790,2508489.0,1896301.7101192577,612187.2898807423
+2022-02-13,False,1426.7739321889594,10233,899.1,2829,2543553.9,1915938.2627990264,627615.6372009735
+2022-02-14,False,1337.9752123091835,10053,899.1,2839,2552544.9,1920840.759277585,631704.140722415
+2022-02-15,True,2351.475059441982,18894,790.8822013542945,6744,5333709.565933362,3874358.9633387243,1459350.6025946373
+2022-02-16,False,1244.9510303702357,9871,899.1,2774,2494103.4,1888252.9386465398,605850.4613534601
+2022-02-17,False,1211.8738687166879,9910,899.1,2711,2437460.1,1856714.7898701457,580745.3101298544
+2022-02-18,False,1291.248092774871,9977,899.1,2701,2428469.1,1851817.548661015,576651.5513389851
+2022-02-19,False,1202.2934128126344,10133,883.1651108098041,2780,2455199.0080512553,1891207.3343598437,563991.6736914115
+2022-02-20,False,1235.804575281713,10084,899.1,2696,2423973.6,1849246.6954483406,574726.9045516595
+2022-02-21,False,1276.1505339787307,10085,899.1,2656,2388009.6,1829286.22085432,558723.37914568
+2022-02-22,False,1437.4003457574522,9904,899.1,2799,2516580.9,1900942.7903318,615638.1096681999
+2022-02-23,False,1472.2042466986154,9964,899.1,2673,2403294.3000000003,1837976.7032677638,565317.5967322364
+2022-02-24,False,1431.896547822394,10172,899.1,2743,2466231.3000000003,1872944.902847496,593286.3971525042
+2022-02-25,False,1310.2798859572426,10076,899.1,2702,2429368.2,1852330.7801247423,577037.4198752579
+2022-02-26,False,1384.62290131771,10043,899.1,2648,2380816.8000000003,1825387.4752716604,555429.3247283399
+2022-02-27,False,1234.0834436842747,9702,899.1,2657,2388908.7,1829743.9021500424,559164.7978499578
+2022-02-28,False,1283.678769021713,9910,899.1,2682,2411386.2,1842284.07082779,569102.1291722101
+2022-03-01,False,1476.1958625388818,10020,899.1,2709,2435661.9,1855990.3574499364,579671.5425500635
+2022-03-02,False,1241.9789919979662,9898,899.1,2682,2411386.2,1842256.0852074374,569130.1147925628
+2022-03-03,False,1169.1942709029515,9892,899.1,2745,2468029.5,1873673.4418196054,594356.0581803946
+2022-03-04,False,1288.837622807191,10030,899.1,2698,2425771.8000000003,1850296.6598978532,575475.1401021471
+2022-03-05,False,1361.0366475166352,10010,899.1,2789,2507589.9,1895865.5077794136,611724.3922205863
+2022-03-06,False,1150.6057039233983,10086,899.1,2773,2493204.3000000003,1887675.9748459037,605528.3251540966
+2022-03-07,False,1148.8405674690443,9990,899.1,2773,2493204.3000000003,1887650.6458073962,605553.6541926041
+2022-03-08,False,1221.47795637576,9850,899.1,2726,2450946.6,1864229.4841102757,586717.1158897243
+2022-03-09,False,1463.0596405593335,10060,899.1,2781,2500397.1,1891973.8578367615,608423.2421632386
+2022-03-10,False,1369.841540397185,10100,899.1,2678,2407789.8000000003,1840391.6702139727,567398.1297860276
+2022-03-11,False,1188.773662552044,9895,899.1,2834,2548049.4,1918191.6567640717,629857.7432359282
+2022-03-12,False,1341.3950242487674,9934,899.1,2660,2391606.0,1831345.182388721,560260.817611279
+2022-03-13,False,1364.0959335247649,10107,899.1,2812,2528269.2,1907376.925745633,620892.2742543672
+2022-03-14,False,1309.221871963138,10024,899.1,2726,2450946.6,1864309.6853946745,586636.9146053256
+2022-03-15,False,1403.7529842149315,10026,899.1,2746,2468928.6,1874411.3054106305,594517.2945893696
+2022-03-16,False,1109.8955954636629,10165,899.1,2802,2519278.2,1902111.2872151928,617166.9127848074
+2022-03-17,False,1176.4468908415815,9992,899.1,2683,2412285.3000000003,1842689.52860202,569595.7713979804
+2022-03-18,False,1486.5486977700477,10155,899.1,2747,2469827.7,1875005.0228379683,594822.6771620319
+2022-03-19,False,1299.096000702733,9897,899.1,2750,2472525.0,1876301.3950001332,596223.6049998668
+2022-03-20,True,2959.005898764264,18781,852.3176505002403,5582,4757637.125092342,3293965.8953605895,1463671.229731752
+2022-03-21,False,1223.2718760871978,9911,899.1,2742,2465332.2,1872229.5640445976,593102.6359554026
+2022-03-22,False,1100.7089544693627,9934,899.1,2697,2424872.7,1849605.1051355922,575267.594864408
+2022-03-23,False,1331.7656967542457,9836,899.1,2683,2412285.3000000003,1842832.6021656336,569452.6978343667
+2022-03-24,False,1499.7913000398273,10077,899.1,2754,2476121.4,1878519.0568463928,597602.3431536071
+2022-03-25,False,1336.0397990084284,10141,899.1,2767,2487809.7,1884850.7572144207,602958.9427855795
+2022-03-26,False,1344.0291715257779,9881,899.1,2807,2523773.7,1904848.162184592,618925.5378154081
+2022-03-27,False,1231.0906450912316,9917,899.1,2755,2477020.5,1878736.1803134407,598284.3196865593
+2022-03-28,False,1194.813018241124,10016,899.1,2799,2516580.9,1900707.1235184686,615873.7764815313
+2022-03-29,False,1303.2626288160477,10058,899.1,2790,2508489.0,1896304.5482847781,612184.4517152219
+2022-03-30,False,1459.914466790956,10093,899.1,2845,2557939.5,1923968.4508365616,633971.0491634384
+2022-03-31,False,1103.9004416458804,9839,899.1,2756,2477919.6,1879111.1779584847,598808.4220415154

docs/source/example_notebooks/do_sampler_demo.ipynb

@@ -45,7 +45,7 @@
     "\n",
     "## Integration\n",
     "\n",
-    "The do-sampler is built on top of the identification abstraction used throughout do-why. It uses a `dowhy.CausalModel` to perform identification, and builds any models it needs automatically using this identification.\n",
+    "The do-sampler is built on top of the identification abstraction used throughout do-why. It automatically performs an identification, and builds any models it needs automatically using this identification.\n",
     "\n",
     "## Specifying Interventions\n",
     "\n",
@@ -128,7 +128,8 @@
     "model = CausalModel(df, \n",
     "                    causes,\n",
     "                    outcomes,\n",
-    "                    common_causes=common_causes)"
+    "                    common_causes=common_causes)\n",
+    "nx_graph = model._graph._graph"
    ]
   },
   {
@@ -162,8 +163,11 @@
    "source": [
     "from dowhy.do_samplers.weighting_sampler import WeightingSampler\n",
     "\n",
-    "sampler = WeightingSampler(df,\n",
-    "                           causal_model=model,\n",
+    "sampler = WeightingSampler(graph=nx_graph,\n",
+    "                           action_nodes=causes,\n",
+    "                           outcome_nodes=outcomes,\n",
+    "                           observed_nodes=df.columns.tolist(),\n",
+    "                           data=df,\n",
     "                           keep_original_treatment=True,\n",
     "                           variable_types={'D': 'b', 'Z': 'c', 'Y': 'c'}\n",
     "                          )\n",
@@ -207,7 +211,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -221,7 +225,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.8.10"
   },
   "toc": {
    "base_numbering": 1,

docs/source/example_notebooks/dowhy_causal_api.ipynb

@@ -16,6 +16,7 @@
    "source": [
     "import dowhy.datasets\n",
     "import dowhy.api\n",
+    "from dowhy.graph import build_graph_from_str\n",
     "\n",
     "import numpy as np\n",
     "import pandas as pd\n",
@@ -36,7 +37,7 @@
     "        treatment_is_binary=True)\n",
     "df = data['df']\n",
     "df['y'] = df['y'] + np.random.normal(size=len(df)) # Adding noise to data. Without noise, the variance in Y|X, Z is zero, and mcmc fails.\n",
-    "#data['dot_graph'] = 'digraph { v ->y;X0-> v;X0-> y;}'\n",
+    "nx_graph = build_graph_from_str(data[\"dot_graph\"])\n",
     "\n",
     "treatment= data[\"treatment_name\"][0]\n",
     "outcome = data[\"outcome_name\"][0]\n",
@@ -47,15 +48,17 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "metadata": {},
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [],
    "source": [
     "# data['df'] is just a regular pandas.DataFrame\n",
     "df.causal.do(x=treatment,\n",
-    "                     variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'},\n",
-    "                     outcome=outcome,\n",
-    "                     common_causes=[common_cause],\n",
-    "                     proceed_when_unidentifiable=True).groupby(treatment).mean().plot(y=outcome, kind='bar')"
+    "             variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'},\n",
+    "             outcome=outcome,\n",
+    "             common_causes=[common_cause],\n",
+    "            ).groupby(treatment).mean().plot(y=outcome, kind='bar')"
    ]
   },
   {
@@ -68,8 +71,8 @@
     "              variable_types={treatment:'b', outcome: 'c', common_cause: 'c'}, \n",
     "              outcome=outcome,\n",
     "              method='weighting', \n",
-    "              common_causes=[common_cause],\n",
-    "              proceed_when_unidentifiable=True).groupby(treatment).mean().plot(y=outcome, kind='bar')"
+    "              common_causes=[common_cause]\n",
+    "              ).groupby(treatment).mean().plot(y=outcome, kind='bar')"
    ]
   },
   {
@@ -81,14 +84,14 @@
     "cdf_1 = df.causal.do(x={treatment: 1}, \n",
     "              variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'}, \n",
     "              outcome=outcome, \n",
-    "              dot_graph=data['dot_graph'],\n",
-    "              proceed_when_unidentifiable=True)\n",
+    "              graph=nx_graph\n",
+    "              )\n",
     "\n",
     "cdf_0 = df.causal.do(x={treatment: 0}, \n",
     "              variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'}, \n",
     "              outcome=outcome, \n",
-    "              dot_graph=data['dot_graph'],\n",
-    "              proceed_when_unidentifiable=True)\n"
+    "              graph=nx_graph\n",
+    "              )\n"
    ]
   },
   {
@@ -158,7 +161,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -172,7 +175,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.8.10"
   },
   "toc": {
    "base_numbering": 1,