-
Notifications
You must be signed in to change notification settings - Fork 0
/
nhosp-gen-models-GRP.py
1448 lines (1024 loc) · 57 KB
/
nhosp-gen-models-GRP.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@author: Carlos Minutti Martinez <carlos.minutti@iimas.unam.mx>
@author: Miguel Félix Mata Rivera <mmatar@ipn.mx >
MODELO DE REGRESIÓN BINOMIAL NEGATIVA PARA PREDECIR EL NÚMERO DE HOSPITALIZACIONES
Resumen:
Mediante el análisis de datos de egresos hospitalarios, se ha desarrollado un
sistema automatizado para la generación de modelos predictivos del número y la
severidad de hospitalizaciones por diversas enfermedades y categorías,
seleccionando los códigos CIE o grupos de códigos a modelar. Para tal fin,
se han empleado técnicas estadísticas, tales como la regresión logística y la
regresión binomial negativa.
No obstante, es importante destacar que los modelos generados no tienen como
finalidad la predicción, ni están destinados a ser utilizados con ese propósito,
sino que su objetivo es interpretar las variables que resultan relevantes a la
hora de examinar la severidad y el número de hospitalizaciones. En este sentido,
los modelos consideran el efecto de múltiples variables, tales como factores
socioeconómicos, exposición a contaminantes ambientales, lugar de residencia,
peso, edad, sexo, densidad de población, derechohabiencia, fecha de ingreso (con
el fin de identificar patrones temporales) y se permite especificar el rango
de años a utilizar. Por lo tanto, el interés radica en comprender cómo estas
variables influyen en el modelo y
la magnitud de su efecto.
Además, para mantener la interpretabilidad del modelo, se incluyen únicamente
aquellas variables que contribuyen de manera significativa al mismo, a fin de
que éste sea lo más parsimonioso posible y se reduzca el problema de
multicolinealidad. Este procedimiento se lleva a cabo a través de un algoritmo
ad-hoc que incluye penalizaciones en el proceso de selección de variables, y
que toma en cuenta tanto la multicolinealidad como la paradoja de Simpson. En
caso de que exista un efecto diferente al esperado (por ejemplo, de un análisis
de correlación), se mantiene únicamente si su contribución al modelo resulta
significativa.
El proceso de generación de los indicadores socioeconómicos y de contaminantes
atmosféricos puede consultarse en los otros resultados de este proyecto.
Cabe destacar que, por motivos de privacidad, la base de datos original no se
comparte. No obstante, se ha generado una versión con datos sintéticos que
permite probar los algoritmos y adaptarlos para su uso en otros proyectos.
Licencia:
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
GitHub: https://github.com/cminuttim/
"""
from IPython import get_ipython;
get_ipython().run_line_magic('reset','-sf')
import pickle
import pandas as pd
import numpy as np
import sqlite3
import matplotlib.pyplot as plt
import statsmodels.api as sm
import statsmodels.formula.api as smf
from tqdm.auto import tqdm
# Configuración del análisis a correr
class CFG:
base_dir = './' # Directorio base
db = 'rnd_db.sqlite' # Base de datos a usar
seed = 11 # Semilla para el componente aleatorio
out = 'out/' # Directorio de salida para las imágenes generadas
save_plot = True # ¿Se guardan las imágenes?
train_prop = 0.85 # Proporción de registros para entrenamiento, (1-train_prop) para validación
grp = 'PC' # Tipo de análisis a realizar, PC: Principales causas de defunción, HE: Hospitalizaciones evitables, CM: Grupo especifico
max_grps = 2 # Número máximo de modelos a generar
use_weight = False # Usar mínimos cuadrados ponderados
lang = 'EN' # lang=EN o lang=ES para la salida de resultados
use_defu = True # ¿Usar solo CIE que tengan defunciones?
q_cond = 'ANIO<=2020' # rango de años a usar para el análisis
# fijar semilla
np.random.seed(CFG.seed)
# CIE
# http://www.dgis.salud.gob.mx/contenidos/intercambio/diagnostico_gobmx.html
# CLUES
# http://www.dgis.salud.gob.mx/contenidos/intercambio/clues_gobmx.html
# PRINCAU
# LISTA MEXICANA PARA LA SELECCION DE LAS PRINCIPALES CAUSASDE MORTALIDAD
# http://dgis.salud.gob.mx/descargas/pdf/lista_mexicana.pdf
### Se cargan los catálogos de CIE Y CLUES ####
con = sqlite3.connect(CFG.base_dir + CFG.db)
cur = con.cursor()
clues = pd.read_sql_query('SELECT * FROM CATCLUES', con)
clues_export = clues.copy()
clues = clues.set_index(clues['CLUES'])
cie = pd.read_sql_query('SELECT * FROM CATCIE10', con)
cie_export = cie.copy()
cie.set_index('CAUSA', inplace=True)
# Catalogo de principales causas de defunción
df_princau = pd.read_sql_query('SELECT PRINCAU.descrip_padre, PRINCAU.principal FROM PRINCAU', con)
cie_caup = pd.read_sql_query('SELECT CATCIE10.CAUSA, PRINCAU.descrip_padre AS PCAU_PADRE_DESC \
FROM CATCIE10 \
INNER JOIN PRINCAU ON PRINCAU.principal = CATCIE10.prinmorta', con)
cie_caup.set_index('CAUSA', inplace=True)
cie = pd.concat([cie, cie_caup], axis=1)
# Catalogo de CIE consideradas cronicas
cie_c = pd.read_csv(CFG.base_dir + 'cie_cronicas.csv')
# Catalogos de hospitalizaciones evitables (Purdy y ACSCMex)
cie_hosp_e_Purdy = pd.read_csv(CFG.base_dir + 'cie-hosp_e_Purdy.csv')
cie_hosp_e_ACSCMex = pd.read_csv(CFG.base_dir + 'cie-hosp_e_ACSCMex.csv')
cie_hosp_e = pd.concat([cie_hosp_e_ACSCMex,cie_hosp_e_Purdy])
cie_hosp_e = cie_hosp_e[['CAUSA', 'CATEGORIA', 'NV0']]
cie_hosp_e.drop_duplicates(inplace=True)
cie_hosp_e.set_index('CAUSA', inplace=True)
cie_hosp_e.dropna(inplace=True)
hosp_e_cat = cie_hosp_e.CATEGORIA.unique()
cie_hosp_gdesc = pd.read_csv(CFG.base_dir + 'cie-hosp_gdesc.csv')
cie_hosp_gdesc.set_index('Code', inplace=True)
con.close()
# Traducción de variables de fecha
day_es = ("Lunes", "Martes", "Miércoles", "Jueves", "Viernes", "Sábado", "Domingo")
day_en = ("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday")
day_en_es = {day_en[i]: day_es[i] for i in range(len(day_en))}
month_es = ("Enero", "Febrero", "Marzo", "Abril", "Mayo", "Junio", "Julio", "Agosto", "Septiembre", "Octubre", "Noviembre", "Diciembre")
month_en = ("January", "February", "March", "April", "May", "June", "July", "August", "September", "October", "November", "December")
month_en_es = {month_en[i]: month_es[i] for i in range(len(month_en))}
# Traducción de algunos grupos de enfermedades
grp_en = {}
grp_en['DC'] = '[E10–E14] Diabetes mellitus'
grp_en['E112'] = '[E112] Type 2 diabetes mellitus with kidney complications'
grp_en['E117'] = '[E117] Type 2 diabetes mellitus with multiple complications'
grp_en['E145'] = '[E145] Unspecified diabetes mellitus: with peripheral circulatory complication'
grp_en['E118'] = '[E118] Type 2 diabetes mellitus with unspecified complications'
grp_en['E10-E149'] = 'Diabetes mellitus'
grp_en['A15-Z999'] = 'Other'
grp_en['J09X-U049'] = 'Influenza and Pneumonia'
grp_en['K70-K769'] = 'Liver diseases'
grp_en['I00X-I519'] = 'Heart diseases'
grp_en['U072-U072'] = 'COVID-19, virus unidentified'
grp_en['I60-I698'] = 'Cerebrovascular Diseases'
grp_en['A33X-P969'] = 'Certain conditions originating in the perinatal period'
grp_en['U071-U071'] = 'COVID-19, virus identified'
grp_en['N17-N19X'] = 'Renal insufficiency'
#########################################
## INEGI VARS
#########################################
#
# Se cargan los datos del Censo de Población y Vivienda 2020 (INEGI), así como
# las variables y ponderaciones para construir los indicadores socioeconómicos por localidad
#
# Valores (999,9999) indican que tienen municipio o localidad desconocida
inegi_vars = pd.read_csv(CFG.base_dir + './variables.csv')
inegi_loc = pd.read_csv(CFG.base_dir + './inegi_loc.csv', dtype={'ENTIDAD':'str','MUN':'str','LOC':'str'})
area_ageb = pd.read_csv(CFG.base_dir + './ageb-area.csv', dtype={'CVE_ENT':'str','CVE_MUN':'str','CVE_LOC':'str','CVE_AGEB':'str'})
inegi_fa_loads = pd.read_csv(CFG.base_dir + './inegi_fa_loads.csv')
scale_inegi_fa = pd.read_csv(CFG.base_dir + './scale_inegi_fa.csv').transpose()
inegi_fa_loads.set_index('Var', inplace=True)
scale_inegi_fa.rename(columns = {0:'min', 1:'max'}, inplace = True)
area_ageb['loc_id'] = area_ageb['CVE_ENT'] + area_ageb['CVE_MUN'] + area_ageb['CVE_LOC']
area_loc = area_ageb.groupby(['loc_id'])['Area'].sum().reset_index()
area_loc.set_index('loc_id', inplace=True)
id_inegi = (inegi_loc['ENTIDAD']+inegi_loc['MUN']+inegi_loc['LOC'])
inegi_loc.set_index(id_inegi, inplace=True)
idx = inegi_loc.index[inegi_loc.index.isin(area_ageb['loc_id'])].to_numpy()
inegi_loc = inegi_loc.assign(AREA=area_loc.Area.min()) #usar el area minima como estimado para aquellos que no aparecen
inegi_loc.loc[idx, 'AREA'] = area_loc.loc[idx, 'Area']
idx_scalable = inegi_loc.columns.isin(inegi_vars.Var[inegi_vars.No_Esc==0])
idx_scale = inegi_loc.columns[idx_scalable]
idx_pob = inegi_loc.columns[[i[0]=='P' for i in inegi_loc.columns]*idx_scalable]
idx_viv = inegi_loc.columns[[i[0]=='V' for i in inegi_loc.columns]*idx_scalable]
inegi_loc_s = inegi_loc.copy()
inegi_loc_s[idx_pob] = inegi_loc[idx_pob].div(inegi_loc['POBTOT'], axis=0)
inegi_loc_s[idx_viv] = inegi_loc[idx_viv].div(inegi_loc['VIVTOT'], axis=0)
inegi_loc_s['POBTOT'] = inegi_loc['POBTOT']/inegi_loc['POBTOT'].sum()
inegi_loc_s['VIVTOT'] = inegi_loc['VIVTOT']/inegi_loc['VIVTOT'].sum()
inegi_loc_s = inegi_loc_s.assign(POB_AREA=inegi_loc['POBTOT']/inegi_loc['AREA'])
id_p = inegi_loc_s.columns.get_loc('POBTOT')
nom_ent = inegi_loc['NOM_LOC']
nom_ent.loc['099990001'] = 'Municipio desconocido'
nom_ent.loc['159990001'] = 'Municipio desconocido'
inegi_loc = inegi_loc_s
inegi_loc.iloc[:,id_p:]
main_inegi_vars = ['POBTOT', 'POB_AREA', 'P_0A2', 'P_18A24', 'P_60YMAS', 'POB0_14', 'POB15_64', 'POB65_MAS', 'REL_H_M', 'PROM_HNV', 'GRAPROES', 'PROM_OCUP', 'P3HLINHE_M', 'P5_HLI_NHE', 'PDER_IMSS', 'PDER_ISTE', 'PDER_ISTEE', 'PAFIL_PDOM', 'PDER_SEGP', 'PDER_IMSSB', 'PAFIL_IPRIV', 'VIVPAR_UT', 'VPH_PISOTI', 'VPH_1CUART', 'VPH_AGUAFV', 'VPH_LETR', 'VPH_NODREN', 'VPH_SNBIEN', 'VPH_SINRTV', 'VPH_SINTIC']
inegi_loc = inegi_loc[main_inegi_vars]
id_inegi_fa = inegi_loc.columns.get_loc('REL_H_M')
id_p = 0
inegi_vars.set_index(inegi_vars.Var, inplace=True)
#--------------------------
#########################################
## CONTAMINANTES
#########################################
#
# Se cargan las concentraciones de contaminantes y las ponderaciones
# para construir los indicadores (factores) de contaminantes por localidad.
# Estos factores tienen la finalidad de solventar el problema de contaminantes
# correlacionados espacialmente (multicolinealidad), al agrupar en factores de
# aquellos que tienden a estar presentes de manera simultanea.
cont_loc = pd.read_csv(CFG.base_dir + './cont_loc_mean.csv', dtype={'ENTIDAD':'str','MUN':'str','LOC':'str'})
id_cont = (cont_loc['ENTIDAD']+cont_loc['MUN']+cont_loc['LOC'])
scale_cont_fa = pd.read_csv(CFG.base_dir + './scale_cont_fa.csv').transpose()
scale_cont_fa.rename(columns = {0:'min', 1:'max'}, inplace = True)
cont_loc.set_index(id_cont, inplace=True)
idx = cont_loc.index.isin(inegi_loc.index)
cont_loc = cont_loc[idx]
cont_loc.drop(['ENTIDAD', 'MUN', 'LOC'], axis=1, inplace=True)
#########################################
## MAIN
#########################################
# carga de datos
db_year = range(15,21)
year = range(2015,2021)
db_file = CFG.base_dir + CFG.db
con = sqlite3.connect(db_file)
cur = con.cursor()
df_afec = pd.read_sql_query(f'SELECT * FROM AFECCIONES WHERE {q_cond}', con)
df_egre = pd.read_sql_query(f'SELECT * FROM EGRESO WHERE ({q_cond} AND (ENTIDAD="09"))', con)
df_defu = pd.read_sql_query(f'SELECT DEFUNC.*, EGRESO.DIAGNOSTICO, EGRESO.AFECCION, EGRESO.DIAS, EGRESO.EDAD, EGRESO.SEXO, EGRESO.PESO, EGRESO.TALLA, EGRESO.PROCED, EGRESO.DERHAB, EGRESO.VEZ, \
EGRESO.ENTIDAD, EGRESO.MUNIC, EGRESO.LOC, EGRESO.INGRESO \
FROM DEFUNC \
INNER JOIN EGRESO on EGRESO.FOLIO = DEFUNC.FOLIO AND EGRESO.CLUES = DEFUNC.CLUES AND EGRESO.EGRESO = DEFUNC.EGRESO \
WHERE (DEFUNC.{q_cond} AND (EGRESO.ENTIDAD="09"))', con)
con.close()
def shuffle_data(df):
n, m = df.shape
df = df.reset_index(drop=True)
for i in range(m):
df.iloc[:,i] = df.iloc[:,i].sample(frac=1).reset_index(drop=True)
return df
# remover espacio al final (array(['0 ', '8 ', 'G ', '9 ', '6 ', '2 '], dtype=object))
df_egre.DERHAB = df_egre.DERHAB.str[:-1]
df_defu.DERHAB = df_defu.DERHAB.str[:-1]
# get summary statistics
num_cols_afec = df_afec._get_numeric_data().columns
df_afec_num_desc = df_afec[num_cols_afec].describe(percentiles=[.05, .25, .5, .75, .95]).transpose()
df_afec_cat_desc = df_afec.iloc[:,~df_afec.columns.isin(num_cols_afec)].describe().transpose()
df_afec_cat_desc['perc'] = (df_afec_cat_desc['freq']/df_afec_cat_desc['count']*100).astype('float')
num_cols_egre = df_egre._get_numeric_data().columns
df_egre_num_desc = df_egre[num_cols_egre].describe(percentiles=[.05, .25, .5, .75, .95]).transpose()
df_egre_cat_desc = df_egre.iloc[:,~df_egre.columns.isin(num_cols_egre)].describe().transpose()
df_egre_cat_desc['perc'] = (df_egre_cat_desc['freq']/df_egre_cat_desc['count']*100).astype('float')
num_cols_defu = df_defu._get_numeric_data().columns
df_defu_num_desc = df_defu[num_cols_defu].describe(percentiles=[.05, .25, .5, .75, .95]).transpose()
df_defu_cat_desc = df_defu.iloc[:,~df_defu.columns.isin(num_cols_defu)].describe().transpose()
df_defu_cat_desc['perc'] = (df_defu_cat_desc['freq']/df_defu_cat_desc['count']*100).astype('float')
# Eliminar outliers
df_egre.loc[df_egre.EDAD>120,'EDAD'] = np.nan
df_egre.loc[df_egre.PESO>250,'PESO'] = np.nan
df_egre.loc[df_egre.TALLA>250,'TALLA'] = np.nan
df_defu.loc[df_defu.EDAD>120,'EDAD'] = np.nan
df_defu.loc[df_defu.PESO>250,'PESO'] = np.nan
df_defu.loc[df_defu.TALLA>250,'TALLA'] = np.nan
# tablas de conteo para los diagnósticos
n_egre_cie = df_egre.groupby('DIAGNOSTICO')['DIAGNOSTICO'].count().sort_values(ascending=False)
n_defu_cie_bas = df_defu[['CAUSA']].groupby('CAUSA')['CAUSA'].count().sort_values(ascending=False)
n_defu_cie_bas = pd.concat([n_defu_cie_bas, cie.loc[n_defu_cie_bas.index,'Nombre']], axis=1)
n_defu_cie_bas.rename(columns = {'CAUSA':'count'}, inplace = True)
# Por edad
edad_cie = df_defu.pivot_table(['EDAD'],
['CAUSA'],
aggfunc=lambda x: [np.percentile(x, [25, 50, 75])],
)
edad_cie = pd.DataFrame(np.asmatrix(list(map(lambda a:a[0],edad_cie.EDAD))),
index=edad_cie.index,
columns=['edad_Q1', 'edad_Q2', 'edad_Q3',])
edad_cie = pd.concat([edad_cie, cie.loc[edad_cie.index,'Nombre']], axis=1)
edad_cie_def = edad_cie.copy()
# Por sexo
sexo_cie = df_defu.pivot_table(['SEXO'],
['CAUSA'],
aggfunc=lambda x: np.round([np.mean(x=='F'), np.mean(x=='M')],3),
)
sexo_cie = pd.DataFrame(np.asmatrix(list(sexo_cie.SEXO)),
index=sexo_cie.index,
columns=['SEXO_F', 'SEXO_M'])
sexo_cie_def = sexo_cie.copy()
idx_sel = edad_cie_def.index.isin(n_defu_cie_bas.index) # should be 100%
cie_def = edad_cie_def.loc[idx_sel,]
cie_def.loc[:,'count'] = n_defu_cie_bas.loc[cie_def.index,'count']
cie_def[['SEXO_F', 'SEXO_M']] = sexo_cie_def.loc[cie_def.index,['SEXO_F', 'SEXO_M']]
# Por edad y sexo
cie_def = cie_def[['edad_Q1', 'edad_Q2', 'edad_Q3', 'SEXO_F', 'SEXO_M', 'Nombre', 'count']].sort_values(by=['count'], ascending=False)
cie_def_sel = cie_def.iloc[:50,]
# Se transforma la fecha a variable de tiempo
df_egre.EGRESO = pd.to_datetime(df_egre.EGRESO)
df_defu.EGRESO = pd.to_datetime(df_defu.EGRESO)
df_afec.EGRESO = pd.to_datetime(df_afec.EGRESO)
# Se construye un identificador único de registro
df_egre['ID'] = df_egre['FOLIO'].map(str) + '_' + df_egre['CLUES'].map(str) + '_' + df_egre['EGRESO'].dt.strftime('%Y-%m-%d')
df_defu['ID'] = df_defu['FOLIO'].map(str) + '_' + df_defu['CLUES'].map(str) + '_' + df_defu['EGRESO'].dt.strftime('%Y-%m-%d')
df_afec['ID'] = df_afec['FOLIO'].map(str) + '_' + df_afec['CLUES'].map(str) + '_' + df_afec['EGRESO'].dt.strftime('%Y-%m-%d')
df_egre.INGRESO = pd.to_datetime(df_egre.INGRESO)
df_defu.INGRESO = pd.to_datetime(df_defu.INGRESO)
# Fecha como un valor numerico del mes-año
df_egre['FECHA'] = df_egre.INGRESO.dt.month + 12*(df_egre.INGRESO.dt.year-2015)
df_defu['FECHA'] = df_defu.INGRESO.dt.month + 12*(df_defu.INGRESO.dt.year-2015)
df_egre['MES'] = df_egre.INGRESO.dt.month_name()
df_defu['MES'] = df_defu.INGRESO.dt.month_name()
df_egre['QN'] = df_egre.INGRESO.dt.quarter
df_defu['QN'] = df_defu.INGRESO.dt.quarter
df_egre['ANIO'] = df_egre.INGRESO.dt.year
df_defu['ANIO'] = df_defu.INGRESO.dt.year
df_egre['DIA_SEMANA'] = df_egre.INGRESO.dt.day_name()
df_defu['DIA_SEMANA'] = df_defu.INGRESO.dt.day_name()
# Función para escalar los datos a 0-1
def min_max_scaler(X, xmin=None, xmax=None):
if type(xmin)==type(None):
xmin = X.min()
else:
xmin = xmin[X.columns.values]
if type(xmax)==type(None):
xmax = X.max()
else:
xmax = xmax[X.columns.values]
X_tmp = X.copy()
# constant values to 0
idx = xmin.index[xmin==xmax]
idx = idx[idx.isin(X_tmp.columns)]
if idx.shape[0]>0:
X_tmp[idx] = 0
idx = xmin.index[(xmin<xmax) * ((xmin!=0) + (xmax!=1))]
idx = idx[idx.isin(X_tmp.columns)] # only available columns
X_tmp[idx] = (X_tmp[idx]-xmin[idx])/(xmax[idx]-xmin[idx])
return (X_tmp, xmin, xmax)
# Función para limitar el rango de una variable
def x_set_lim(X, ll, ul):
X_l = X.copy()
X_l[X_l<ll] = ll
X_l[X_l>ul] = ul
return X_l
#------------------------------------------------------------------
# Modelos
#------------------------------------------------------------------
df_egre_not_defu = df_egre.loc[~df_egre['ID'].isin(df_defu['ID']),]
# lista de códigos CIE para las defunciones en la base de datos
causas_defu = pd.Series(df_defu.CAUSA.unique())
cie_xcat = causas_defu[~(causas_defu.isin(df_egre_not_defu.AFECCION) + causas_defu.isin(df_egre_not_defu.DIAGNOSTICO))]
cie_xcat = cie_xcat[cie_xcat.isin(cie_def.index)].values
cie_def.drop(cie_xcat, axis=0, inplace=True)
# percentage of death causes in the chronic db
idx_c = cie_def.index.isin(cie_c['cie'])
idx_c.sum()/idx_c.shape[0]
cie_def_c = cie_def.iloc[idx_c,]
np.log10(cie_def['count']).hist()
n_egre_in_def_cie = n_egre_cie[n_egre_cie.index.isin(cie_def.index)]
# al menos 20 defunciones con esa causa (por posibles CIE que se pusieron como causa de del pero no lo son)
n_egre_in_def_cie = n_egre_in_def_cie[cie_def['count'][n_egre_in_def_cie.index]>20]
# Inicialización de variables para los modelos a generar
cum_val_sc_reg = 0
cum_val_sc_reg2 = 0
cum_val_sc_reg3 = 0
cum_val_sc_tree = 0
cum_val_sc_gbm = 0
cum_val_sc_hgbm = 0
cum_val_mse = 0
cum_val_mse2 = 0
cum_val_sc_mix = 0
models_dict = {}
models = []
model_vars = pd.Series(dtype=object)
# Calcular la variable dependiente Y
# Usa días de hospitalización como variable de severidad y afecciones
# relaciona menos días en defunciones con más severidad
# menos días en egresos (no defunc) con menos severidad
# avoid warnings using assign
df_egre_not_defu = df_egre_not_defu.assign(Y=0.0)
df_defu = df_defu.assign(Y=1.0)
# Create Y as 0-1
max_d = 150
fx = np.log(max_d+1)-np.log(df_defu.DIAS+1)
fx[fx<0] = 0
df_defu['Y'] = 1/(1+np.exp(-1.0*fx))
fx = np.log(df_egre_not_defu.DIAS+1)-np.log(max_d+1)
fx[fx>0] = 0
df_egre_not_defu['Y'] = 1/(1+np.exp(-1.0*fx))
# Variable para almacenar cuantas veces fue relevante una variable socioeconómica en los modelos
inegi_v_model = pd.DataFrame(['F_ECONOM', 'F_SOCIAL'], columns=['var'])
inegi_v_model = inegi_v_model.assign(count=0)
inegi_v_model.set_index('var', inplace=True)
# Variable para almacenar cuantas veces fue relevante una variable de contaminantes en los modelos
cont_v_model = pd.DataFrame(['PM_CO', 'NO2_NOx', 'SO2_NO_O3'], columns=['var'])
cont_v_model = cont_v_model.assign(count=0)
cont_v_model.set_index('var', inplace=True)
# lista de localidades
loc_id = pd.DataFrame(df_defu['ENTIDAD'] + df_defu['MUNIC'] + df_defu['LOC'], columns=['loc'])
# registros por localidad
loc_n = loc_id.groupby(['loc'])['loc'].count().sort_values(ascending = False)
prop_ine = []
# Inicialización de variables
cie_id = []
cie_name = []
cie_reg_err = []
cie_reg_cor = []
cie_inegi_vars = []
cie_cont_vars = []
cie_emun_vars = []
cie_negv = []
cie_time = []
cie_sexo = []
cie_only_pobtot = []
cie_cronic = []
cie_pred_outliers = []
cie_n_cases = []
# Determinar la lista de códigos CIE o de grupos a utilizar según la configuración
cie_grp = {}
cie_grp_desc = {}
grp_n_regs = {}
if CFG.grp == 'HE':
for i,grp in enumerate(hosp_e_cat):
cie_grp[grp] = cie_hosp_e.index[cie_hosp_e.CATEGORIA == grp].tolist()
cie_grp_desc[grp] = cie_hosp_gdesc.ES_Desc[grp]
cie_tmp = pd.Series(cie_grp[grp])
cie_tmp = cie_tmp[cie_tmp.str.len()==3]
for j,x in enumerate(cie_tmp):
cie_grp[grp] += cie.index[cie.index.str.match(x+'.')].tolist()
grp_n_regs[grp] = ((df_egre.AFECCION.isin(cie_grp[grp])) + (df_egre.DIAGNOSTICO.isin(cie_grp[grp]))).sum()
if CFG.grp == 'PC':
pcau_padre = cie.PCAU_PADRE_DESC.dropna().unique()
for i,grp in enumerate(pcau_padre):
cie_lst = cie.index[cie.PCAU_PADRE_DESC == grp].tolist()
grp_id = cie_lst[0]+'-'+cie_lst[-1]
cie_grp[grp_id] = cie_lst
cie_grp_desc[grp_id] = grp
grp_n_regs[grp_id] = ((df_egre.AFECCION.isin(cie_lst)) + (df_egre.DIAGNOSTICO.isin(cie_lst))).sum()
if CFG.use_defu:
grp_n_regs[grp_id] = df_defu.CAUSA.isin(cie_lst).sum()
if CFG.grp == 'CM':
cie_lst = cie.index[cie.PCAU_PADRE_DESC == 'Diabetes mellitus'].tolist()
idx = n_egre_in_def_cie.index.isin(cie_lst)
idx_cie = n_egre_in_def_cie.index[idx]
for i,grp in enumerate(idx_cie):
cie_grp[grp] = [grp]
cie_grp_desc[grp] = cie.Nombre[grp]
grp_n_regs[grp] = ((df_egre.AFECCION.isin([grp])) + (df_egre.DIAGNOSTICO.isin([grp]))).sum()
df_grp = pd.DataFrame(data={'grp':cie_grp.keys(), 'cie':cie_grp.values(), 'desc':cie_grp_desc.values(), 'n_regs':grp_n_regs.values()})
df_grp = df_grp.sort_values(by=['n_regs'], ascending=False).reset_index(drop=True)
df_grp.set_index('grp', inplace=True)
print(df_grp)
df_princau = df_princau[df_princau.descrip_padre.isin(df_grp.desc.values[1:20])]
def rep_outlier(X, d=0.001, max_p=1.4):
X_up = X.quantile(1-d)
idx_up = X.max().div(X_up)>max_p
X_c = X.copy()
for i in X.columns[idx_up]:
X_c.loc[X_c[i]>X_up[i],i] = X_up[i]
return X_c
gbm_score_dict = {}
#------------------------------------------------------------------
# cie list
#------------------------------------------------------------------
for cie_iter, grp in enumerate(tqdm(df_grp.index[:CFG.max_grps], desc='Modelo')):
# Codigos CIE y descripcion para el grupo actual
cie_x = df_grp.cie[grp]
cie_x_desc = df_grp.desc[grp]
# Determinar la lista de registros que cumplen con el grupo a estudiar
df_egre_cie = df_egre_not_defu.loc[(df_egre_not_defu.AFECCION.isin(cie_x)) + (df_egre_not_defu.DIAGNOSTICO.isin(cie_x)), :]
df_defu_cie = df_defu.loc[(df_defu.AFECCION.isin(cie_x)) + (df_defu.DIAGNOSTICO.isin(cie_x)) + (df_defu.CAUSA.isin(cie_x)), :]
id_lst = pd.concat([df_egre_cie.ID, df_defu_cie.ID])
tmp = []
# Se reinicia semilla para cada grupo a estudiar
np.random.seed(CFG.seed)
#------------------------------------------------------------------
# Data Frame
#------------------------------------------------------------------
vars_n = ['Y', 'EDAD', 'SEXO', 'PESO', 'PROCED', 'VEZ', 'ENTIDAD', 'MUNIC', 'LOC', 'FECHA', 'MES', 'ANIO']
tmp = pd.concat([df_egre_cie[vars_n], df_defu_cie[vars_n]], axis=0).set_index(id_lst)
X_df = tmp
X_df.isnull().sum()
X_df['EDAD'].fillna(X_df['EDAD'].median(), inplace = True)
X_df['PESO'].fillna(X_df['PESO'].median(), inplace = True)
# La variable TALLA no es confiable, por lo que no se usa
#X_df['TALLA'].fillna(X_df['TALLA'].median(), inplace = True)
#------------- VARIABLES SOCIOECONOMICAS ------------------
loc_id = pd.DataFrame(X_df['ENTIDAD'] + X_df['MUNIC'] + X_df['LOC'], columns=['loc'])
idx_avail_loc = loc_id['loc'].isin(inegi_loc.index)
prop_ine.append(idx_avail_loc.mean())
X_df = X_df[idx_avail_loc]
loc_id = loc_id[idx_avail_loc]
X_inegi = inegi_loc.loc[loc_id['loc'],:]
X_inegi = X_inegi.iloc[:,id_p:]
X_inegi.set_index(X_df.index, inplace=True)
X_inegi_s, X_inegi_min, X_inegi_max = min_max_scaler(X_inegi.iloc[:,id_inegi_fa:], scale_inegi_fa['min'], scale_inegi_fa['max'])
X_inegi_s = x_set_lim(X_inegi_s, -1, 10) # avoid outliers
X_inegi_fa = X_inegi_s.dot(inegi_fa_loads.loc[X_inegi_s.columns,:])
X_inegi_fa.rename(columns = {'F1':'F_ECONOM', 'F2':'F_SOCIAL'}, inplace = True)
X_df = pd.concat([X_df, X_inegi_fa, X_inegi[['POBTOT', 'POB0_14', 'POB15_64', 'POB65_MAS', 'REL_H_M', 'POB_AREA']]], axis=1)
X_df['E_MUN'] = X_df['ENTIDAD'] + X_df['MUNIC']
X_df.drop(['MUNIC', 'LOC'], axis=1, inplace=True)
#-------------------
X_df['MES_ANIO_LOC'] = X_df['MES'] + '_' + X_df['ANIO'].astype(str) + '_' + loc_id['loc']
#------------- VARIABLES DE CONTAMINANTES ------------------
idx_avail_loc = loc_id['loc'].isin(cont_loc.index)
# Mantener unicamente localidades conocidas
X_cont = cont_loc.loc[loc_id['loc'],:]
X_cont.set_index(X_df.index, inplace=True)
X_cont_s, X_cont_min, X_cont_max = min_max_scaler(X_cont, scale_cont_fa['min'], scale_cont_fa['max'])
X_cont_s['PM_CO'] = 0.35*X_cont_s['pm10_mean'] + 0.39*X_cont_s['pm25_mean'] + 0.26*X_cont_s['co_mean']
X_cont_s['NO2_NOx'] = 0.54*X_cont_s['no2_mean'] + 0.46*X_cont_s['nox_mean']
X_cont_s['SO2_NO_O3'] = 0.35*X_cont_s['so2_mean'] + 0.33*X_cont_s['no_mean'] + 0.32*X_cont_s['o3_mean']
X_cont_s.drop(cont_loc.columns, axis=1, inplace=True)
X_df = pd.concat([X_df, X_cont_s], axis=1)
#-------------------
# Predictoras
X = X_df.copy()
# Incluir otras variables categóricas a través de dummies
X = pd.concat((X, pd.get_dummies(X['SEXO'], prefix='SEXO', drop_first=True)), axis=1)
X = pd.concat((X, pd.get_dummies(X['E_MUN'], prefix='E_MUN', drop_first=False)), axis=1)
X = pd.concat((X, pd.get_dummies(X['ENTIDAD'], prefix='ENTIDAD', drop_first=True)), axis=1)
X = pd.concat((X, pd.get_dummies(X['MES'], prefix='MES', drop_first=False)), axis=1)
y = X['Y']
X.drop(['SEXO', 'PROCED', 'VEZ'], axis=1, inplace=True)
X.drop(['ENTIDAD', 'E_MUN', 'MES', 'ANIO'], axis=1, inplace=True)
X_df.drop(['ENTIDAD'], axis=1, inplace=True)
bool_cols = X.columns[X.dtypes=='bool']
X[bool_cols] = X[bool_cols].astype('int')
X.describe().transpose()
X.dtypes.unique()
X.reset_index(drop=True, inplace=True)
y.reset_index(drop=True, inplace=True)
# Eliminar algunos outliers encontrados y errores de captura
if any((X.EDAD==0) * (X.PESO>10)):
X.loc[(X.EDAD==0) * (X.PESO>10),'PESO'] = X.PESO[X.EDAD==0].median()
if np.quantile(X.EDAD, 0.995)==0 and X.EDAD.max()>0: # Para cuando la mayoria son edad 0
X.loc[X.EDAD>0,'EDAD'] = 0
data_err1 = (X.EDAD>10) * (X.PESO<10)
data_err2 = (X.PESO==9.999) + (X.PESO==9.099) + (X.PESO==0.999)
if any(data_err1): # Pesos sin sentido por culpa de variaviones de 999
X.loc[data_err1,'PESO'] = X.PESO[~data_err1].median()
if any(data_err2): # variaciones de 999, como 9, 9.99, 0.99 u otros para pesos no especificados
X.loc[data_err2,'PESO'] = X.PESO[~data_err2].median()
# Incluir contaminantes, socioeconómicas
main_var = pd.Series(['EDAD', 'PESO', 'SEXO_M', 'F_ECONOM', 'F_SOCIAL', 'PM_CO', 'NO2_NOx', 'SO2_NO_O3'])
main_var = main_var[main_var.isin(X.columns)] # Algunas como SEXO_M podrian no existir por ser todas mujeres
X_main = X[main_var]
X.drop(['EDAD', 'PESO', 'SEXO_M'], axis=1, inplace=True, errors='ignore')
# -------------------------------------------------
# ----- X y Y por mes, se usa media o mediana, y se puede eliminar o no valores 0
# ----- Y es el número ce casos por mes y localidad
# -------------------------------------------------
y = X.groupby('MES_ANIO_LOC')['MES_ANIO_LOC'].count()
X = X.groupby('MES_ANIO_LOC').mean()
X['Y'] = y
# Reescalar los predictores
X_s, xmin, xmax = min_max_scaler(X)
X_s_desc = X_s.describe().transpose()
X_s['Y'] = X['Y']
y_corr = X_s.corr(method='spearman')
corr_y = y_corr[['Y']][1:]
best_vars = y_corr['Y'].abs().sort_values(ascending=False).index
msk = np.random.rand(len(X_s)) < CFG.train_prop
train_desc = X_s[msk].describe().transpose()
test_desc = X_s[~msk].describe().transpose()
train_std = X_s.columns[train_desc['std']>0]
X_s = X_s[train_std]
X = X[train_std]