/
05.1_commute_scenario_building.do
1026 lines (937 loc) · 70.8 KB
/
05.1_commute_scenario_building.do
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
clear
clear matrix
cd "I:\Github_Maxtor"
x
/***********************
* PREPARE THE INPUT INDIVIDUAL-LEVEL SYNTHETIC POP FROM CENSUS [overlap with metahit work] (INPUT 1)
***********************
** RAW INPUTS TO MAKE 'Census2011EW_AllAdults.dta' [all safe-guarded]
* LSOA-level Age*Sex*T2!: WICID WM12EW[CT0489]_lsoa.csv
* Ethnicity, all flows: WICID wu08cew_msoa_v1.csv
* Ethnicity * T2W, selected flows: WICID CT0600.xls --> use to make CT0600_edit.xls editting top rows
* Car ownership, all flows: WICID wu09buk_msoa_v1.csv
* Car ownership * T2W, selected flows: WICID CT0599.xls --> use to make CT0599_edit.xls editting top rows
* Individual-level data 5% sample, regional level: UKDA, 'isg_regionv2.dta' from https://discover.ukdataservice.ac.uk/catalogue/?sn=7605&type=Data%20catalogue
* Individual-level data 5% sample, LA level: UKDA, 'recodev12.dta' from https://discover.ukdataservice.ac.uk/catalogue/?sn=7682&type=Data%20catalogue
* Area-level Index Multiple Deprivation 2015 (Eng)/2014 (Wales)
* Area-level Rural Urban Classification 2011
** DO FILES TO MAKE 'Census2011EW_AllAdults.dta':
* "..\1 - Phys Act_1-PA main\2017_MetaHIT_analysis\2_program\0.1a_CensusCommuteSP_build_180306.do"
* "..\1 - Phys Act_1-PA main\2017_MetaHIT_analysis\2_program\0.1b_CensusNonCommuteSP_build_180306.do"
***********************
* PREPARE GEOG LOOK UP FILES
***********************
import delimited "pct-inputs\02_intermediate\x_temporary_files\unzip\OA11_LSOA11_MSOA11_LAD11_EW_LUv2.csv", varnames(1) clear
drop oa11cd
duplicates drop
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\LSOA11_MSOA11_LAD11_EW_LUv2.dta", replace
foreach x in msoa lsoa {
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\LSOA11_MSOA11_LAD11_EW_LUv2.dta", clear
rename lso11anm lsoa11nm
rename `x'11cd geo_code
rename `x'11nm geo_name
rename lad11nm lad_name
keep geo_code geo_name lad11cd lad_name
duplicates drop
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\`x'\geo_code_lookup.dta", replace
}
***************************
** MORTALITY DATA FOR COMMUTERS (INPUT 2)
***************************
import excel "pct-inputs\01_raw\03_health_data\commute_microsim\LA_WeightedMortality16plus_2016.xls", sheet("Commuters_MortRate") clear firstrow
* FIX GEOGRAPHIES
gen temp=1
recode temp 1=2 if geography=="Cornwall,Isles of Scilly"
recode temp 1=2 if geography=="Westminster,City of London"
expand temp
bysort geography: gen littlen=_n
* FIX PLACES WITH SMALL AREAS MERGED
replace geographycode="E06000053" if geographycode=="E06000052" & littlen==2 // ISLES OF SCILLY, NOT CORNWALL
replace geography="Cornwall" if geographycode=="E06000052"
replace geography="Isles of Scilly" if geographycode=="E06000053"
replace geographycode="E09000001" if geographycode=="E09000033" & littlen==2 // CITY OF LONDON, NOT WESTMINSTER
replace geography="Westminster" if geographycode=="E09000033"
replace geography="City of London" if geographycode=="E09000001"
* FIX PLACES WITH 2014 NOT 2011 LA CODES
replace geographycode="E06000048" if geographycode=="E06000057"
replace geographycode="E07000100" if geographycode=="E07000240"
replace geographycode="E07000104" if geographycode=="E07000241"
replace geographycode="E07000097" if geographycode=="E07000242"
replace geographycode="E07000101" if geographycode=="E07000243"
replace geographycode="E08000020" if geographycode=="E08000037"
drop temp littlen
* RENAME
rename *16to24* *age1*
rename *25to34* *age2*
rename *35to49* *age3*
rename *50to64* *age4*
rename *65to74* *age5*
rename *75plus* *age6*
rename *fem* *sex2*
rename *mal* *sex1*
rename mort* mortrate1*
rename *_sex* *0*
rename *_age* *0*
* RESHAPE
gen littlen=_n
reshape long mortrate,i(littlen) j(celltype)
tostring celltype, gen(typestring)
gen sex=substr(typestring,3,1)
gen agecat=substr(typestring,5,1)
destring sex agecat, replace
gen female=sex-1
rename geographycode home_lad11cd
order home_lad11cd female agecat mortrate
keep home_lad11cd-mortrate
compress
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\mortrate_individual_CommuteSP.dta", replace
**********************************
* PREPARE CYCLE STREETS DATA - SAVE AS STATA (INPUT 3)
**********************************
import delimited "pct-inputs\02_intermediate\02_travel_data\commute\msoa\rfrq_all_data.csv", clear
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\msoa\rfrq_all_data.dta", replace
import delimited "pct-inputs\02_intermediate\02_travel_data\commute\lsoa\rfrq_all_data.csv", clear
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lsoa\rfrq_all_data.dta", replace
****************************
* PREPARE APS SICK HOURS / YEAR
****************************
use "pct-inputs\01_raw\04_other_data\APS_2016-2018\stata\stata11\aps_3yr_jan16dec18_eul.dta", clear
rename *, lower
** SELECT SAMPLE
drop if gor9d=="N99999999" | gor9d=="S99999999" // Eng + Wales
keep if ilodefr==1 // in work
keep if home==3 | home==4 // commuters
keep if illwk!=-9 & home!=-9 & tothrs!=-9 & ttushr>=0 & tothrs>=0 // has all needed data
** PREPARE AGE/SEX/REGION
recode age 16/24=1 25/34=2 35/49=3 50/64=4 65/74=5 75/max=5, gen(agecat5) // over 75 recoded as too small
recode sex 1=0 2=1, gen(female)
gen home_gor=substr(gor9d,-2,2)
destring home_gor, replace
recode home_gor 99=10
** GENERATE HOURS LOST TO SICKNESS AND TOTAL HOURS
* https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/labourproductivity/articles/sicknessabsenceinthelabourmarket/2014-02-25
gen hrsoffsick=ttushr - ttachr
replace hrsoffsick=0 if yless6!=6
gen tothours=ttachr
replace tothours=ttushr if yless6==6
/** FOOTNOTE IN TEXT ON FEMALES
bysort female: egen npeople=sum(pwta18c)
bysort female: egen tot_hrsoffsick=sum(hrsoffsick*pwta18c*52)
bysort female: egen tot_tothours=sum(tothours*pwta18c*52)
gen sickness_rate = tot_hrsoffsick*100/tot_tothours
gen hours_year=tot_tothours/npeople
gen sickness_hours_year=tot_hrsoffsick/npeople
keep female hours_year sickness_rate sickness_hours_year
duplicates drop
list
*/
** GENERATE SICKNESS HOURS PER YEAR AND SICKNESS RATE AVERAGES BY GROUP
bysort female agecat5 home_gor : egen npeople=sum(pwta18c)
bysort female agecat5 home_gor: egen tot_hrsoffsick=sum(hrsoffsick*pwta18c*52/1000000)
bysort female agecat5 home_gor: egen tot_tothours=sum(tothours*pwta18c*52/1000000)
gen sickness_rate = tot_hrsoffsick*100/tot_tothours
gen sickness_hours_year=tot_hrsoffsick*1000000/npeople
keep npeople female agecat5 home_gor tot_hrsoffsick sickness_rate sickness_hours_year
duplicates drop
/* close though not identical (diff years/subsample) to https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/labourproductivity/articles/sicknessabsenceinthelabourmarket/2016
ta female [fw=npeople], sum(sickness_rate)
ta agecat5 [fw=npeople], sum(sickness_rate)
ta home_gor [fw=npeople], sum(sickness_rate)
*/
keep female agecat5 home_gor sickness_hours_year
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_sickness_hours_year.dta", replace version(12)
**********************************
* IMPORT PREPARED REGIONAL SALARY DATA - SAVE AS STATA
**********************************
import excel "pct-inputs\01_raw\04_other_data\Salaries\earnings by region.xlsx", sheet("TAG_scaled_salary") cellrange(E1:H11) firstrow clear
destring *, replace
keep home_gor salary_hourly
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_salary_hourly.dta", replace
**********************************
* IMPORT PREPARED HILLINESS MMET/SPEED DATA - SAVE AS STATA
**********************************
import excel "pct-inputs\02_intermediate\03_hilliness_calculations\EngWales_mmetspeed_hilliness.xlsx", sheet("Hilliness_pct-scripts_input") firstrow clear
destring *, replace
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_mmetspeed_hilliness.dta", replace
*/
*****************
** MERGE OD DATA
*****************
use "..\1 - Phys Act_1-PA main\2017_MetaHIT_analysis\1b_datacreated\Census2011EW_AllAdults.dta", clear
drop census_id ecactivity commute_distcat commute_bicycle12- commute_bus12
* DROP WORK FROM HOME OR NON-COMMUTERS
drop if work_lsoa=="OD0000001" | work_lsoa==""
* MERGE IN MORT RATES
merge m:1 home_lad11cd agecat female using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\mortrate_individual_CommuteSP.dta", nogen
* MERGE IN CYCLE STREETS VARIABLES
rename home_lsoa geo_code1
rename work_lsoa geo_code2
merge m:1 geo_code1 geo_code2 using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lsoa\rfrq_all_data.dta", keepus(rf_dist_km rf_avslope_perc)
drop if _merge==2
drop _merge
rename geo_code1 geocodetemp
rename geo_code2 geo_code1
rename geocodetemp geo_code2
merge m:1 geo_code1 geo_code2 using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lsoa\rfrq_all_data.dta", keepus(rf_dist_km rf_avslope_perc) update
drop if _merge==2
drop _merge
rename geo_code1 work_lsoa
rename geo_code2 home_lsoa
*****************
** STEP 1: GENERATE FLOW TYPE
*****************
gen work_lsoatemp=substr(work_lsoa,1,1)
replace work_lsoa="OutsideEW" if work_lsoatemp=="S" | work_lsoatemp=="N" | work_lsoa=="OD0000002" | work_lsoa=="OD0000004"
* Make 'other' if S = scotland, N = northern ireland, OD0000002 = offshore installation, OD0000004 = otherwise overseas
gen flowtype=.
recode flowtype .=1 if (work_lsoatemp=="E" | work_lsoatemp=="W") & rf_dist_km!=. & home_lsoa!=work_lsoa
recode flowtype .=2 if home_lsoa==work_lsoa
recode flowtype .=3 if work_lsoa=="OD0000003"
recode flowtype .=4 if (work_lsoatemp=="E" | work_lsoatemp=="W") & rf_dist_km==.
recode flowtype .=4 if work_lsoa=="OutsideEW"
label def flowtypelab 1 "England/Wales between-zone" 2 "Within zone" 3 "No fixed place" 4 "Over max dist or outside England/Wales, offshore installation", modify
label val flowtype flowtypelab
drop work_lsoatemp
replace work_lsoa="Other" if flowtype==4
replace work_msoa="Other" if flowtype==4
/** COUNT NO PEOPLE PER FLOWTYPE - TABLE 1 IN MANUAL C1
ta flowtype // % commuters
ta flowtype if commute_mainmode9==1 // % cyclists
*/
*****************
** STEP 2: ASSIGN VALUES IF DISTANCE/HILLINESS UNKNOWN
*****************
** ASSIGN ESTIMATED HILLINESS + DISTANCE VALUES IF START AND END IN SAME PLACE
by home_lsoa work_lsoa, sort: gen littlen1=_n
by littlen1 home_lsoa (rf_dist_km), sort: gen littlen2=_n
foreach x in rf_dist_km rf_avslope_perc {
gen `x'temp=`x'
replace `x'temp=. if littlen2>3 | littlen1>1
bysort home_lsoa: egen `x'temp2=mean(`x'temp)
replace `x'=`x'temp2 if flowtype==2 & `x'==.
}
replace rf_dist_km=rf_dist_km/3 if flowtype==2
* DISTANCE = A THIRD OF THE MEAN DISTANCE OF SHORTEST 3 FLOWS
* HILLINESS = MEAN HILLINESS OF SHORTEST 3 FLOWS
recode rf_dist_km .=0.79 if (home_lsoa=="E02006781" | home_lsoa=="E01019077") & home_lsoa==work_lsoa //ISLES OF SCILLY [msoa/lsoa] - ESTIMATED FROM DISTANCE DISTRIBUTION
recode rf_avslope_perc .=0.2 if (home_lsoa=="E02006781" | home_lsoa=="E01019077") & home_lsoa==work_lsoa //ISLES OF SCILLY [msoa/lsoa] - ESTIMATED FROM CYCLE STREET
drop littlen1 littlen2 rf_dist_kmtemp rf_dist_kmtemp2 rf_avslope_perctemp rf_avslope_perctemp2
** ASSIGN DISTANCE + HILLINESS *AMONG CYCLISTS* VALUES IF NO FIXED PLACE : MEAN DIST AMONG CYCLISTS TRAVELLING <15KM
foreach x in 15 30 {
gen bicycle`x'=(commute_mainmode9==1)
replace bicycle`x'=. if rf_dist_km>`x' | flowtype>2
}
foreach var in dist_km avslope_perc {
gen cyc_`var'=rf_`var'
foreach x in 15 30 {
gen rf_`var'`x'=rf_`var'
replace rf_`var'`x'=. if bicycle`x'==.
}
bysort home_lsoa: egen numnrf_`var'15=sum(rf_`var'15*bicycle15)
bysort home_lsoa: egen denrf_`var'15=sum(bicycle15)
gen meanrf_`var'15=numnrf_`var'15/denrf_`var'15
replace cyc_`var'=meanrf_`var'15 if flowtype==3
}
** ASSIGN DISTANCE + HILLINESS *AMONG CYCLISTS* VALUES IF OUTSIDE ENG/WALES OR >30KM: MEAN DIST AMONG CYCLISTS TRAVELLING <30KM
foreach var in dist_km avslope_perc {
egen numnrf_`var'30=sum(rf_`var'30*bicycle30)
egen denrf_`var'30=sum(bicycle30)
gen meanrf_`var'30=numnrf_`var'30/denrf_`var'30
replace cyc_`var'=meanrf_`var'30 if flowtype==4
replace cyc_`var'=meanrf_`var'30 if flowtype==3 & cyc_`var'==. // USE 30KM if NO LOCAL CYCLIST GOING <15KM
}
order home_lsoa- work_msoa flowtype rf_dist_km rf_avslope_perc cyc_dist_km cyc_avslope_perc mortrate commute_mainmode9 female agecat nonwhite nocar urbancat5 sparse incomedecile
keep home_lsoa-incomedecile
compress
saveold "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\CommuteSPTemp1.dta", replace
*****************
** STEP 3A: CALCULATE PROPENSITY TO CYCLE
*****************
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\CommuteSPTemp1.dta", clear
** INPUT PARAMETERS FLOW TYPE 1+2
recode commute_mainmode9 2/max=0,gen(bicycle)
recode agecat 6=5, gen(agecat5)
recode agecat 1/3=.,gen(agecat_old)
recode agecat 1/3=1 4/6=2, gen(femagecat)
replace femagecat=(femagecat+10) if female==1
label def femagecatlab 1 "Male 16 to 49" 2 "Male 50+" 11 "Female 16 to 49" 12 "Female 50+"
label val femagecat femagecatlab
recode incomedecile 1/2=1 3/4=2 5/6=3 7/8=4 9/10=5,gen(incomefifth)
recode urbancat5 1/2=. ,gen(urbancat5_3only)
gen rf_dist_kmsq=rf_dist_km^2
gen rf_dist_kmsqrt=sqrt(rf_dist_km)
gen ned_rf_avslope_perc=rf_avslope_perc-0.78 // CENTRE ON 13TH PERCENTILE, AS DUTCH AVERAGE HILLINESS EQUAL TO 13TH PERCENTILE AREAS ENGLAND
gen interact=rf_dist_km*ned_rf_avslope_perc
gen interactsqrt=rf_dist_kmsqrt*ned_rf_avslope_perc
** FIT REGRESSION EQUATION - FLOW TYPE 1+2 - GOV TARGET, DUTCH, EBIKE [model derived from MSOA trips]
*xi: logit bicycle rf_dist_km rf_dist_kmsqrt rf_dist_kmsq ned_rf_avslope_perc interact interactsqrt if flowtype <=2
gen pred_basegt= /*
*/ -4.018 + (-0.6369 * rf_dist_km) + (1.988 * rf_dist_kmsqrt) + (0.008775 * rf_dist_kmsq) + (-0.2555 * ned_rf_avslope_perc) + (0.02006 * rf_dist_km*ned_rf_avslope_perc) + (-0.1234 * rf_dist_kmsqrt*ned_rf_avslope_perc)
gen bdutch = 2.550+(-0.08036*rf_dist_km) // FROM DUTCH NTS [could in future do this by age/sex? each time comparing total Eng/Wales pop to a specific Dutch pop]
replace bdutch=. if flowtype==3
gen bebike= (0.05509*rf_dist_km)+(-0.000295*rf_dist_kmsq) // PARAMETERISED FROM DUTCH TRAVEL SURVEY, ASSUMING NO DIFFERENCE AT 0 DISTANCE
replace bebike=bebike+(0.1812 *ned_rf_avslope_perc) // SWISS TRAVEL SURVEY
gen pred_dutch= pred_basegt + bdutch
gen pred_ebike= pred_dutch + bebike
/** NEAR MARKET MODELLING DECISIONS: DIST DECAY BY AGE AND GENDER: FIGURE 2 IN C2 AND ALSO APPENDIX 2 FIGURE/NUM IN TEXT
gen rf_dist_kmround2=(floor(rf_dist_km/2))*2
gen rf_avslope_percround2=floor(rf_avslope_perc)
recode rf_avslope_percround2 6/max=6 // top 1%
count if flowtype<=2
table rf_dist_kmround2 femagecat if flowtype<=2 , c(mean bicycle)
table rf_avslope_percround2 femagecat if flowtype<=2 , c(mean bicycle)
table home_gordet nocar , c(mean bicycle)
*/
/** IDENTIFY REGRESSION EQUATION - FLOW TYPE 1+2 - NEAR MARKET. Paste to 'pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\1logfiles\NearMarket_flow1+2.xls'
log using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\1logfiles\NearMarket_flow1+2.smcl", replace
foreach x in 1 11 {
forval i=1/9 {
disp "femagecat `x' in region `i'"
xi: logit bicycle i.agecat nonwhite nocar i.incomefifth i.urbancat5 sparse rf_dist_km rf_dist_kmsqrt rf_dist_kmsq ned_rf_avslope_perc interact interactsqrt /*
*/ if femagecat==`x' & home_gordet==`i' & flowtype <=2
}
forval i=10/11 { // SW and Wales = only larger cities
disp "femagecat `x' in region `i'"
xi: logit bicycle i.agecat nonwhite nocar i.incomefifth i.urbancat5_3only sparse rf_dist_km rf_dist_kmsqrt rf_dist_kmsq ned_rf_avslope_perc interact interactsqrt /*
*/ if femagecat==`x' & home_gordet==`i' & flowtype <=2
}
}
foreach x in 2 12 { // older ages = agecat_old
forval i=1/9 {
disp "femagecat `x' in region `i'"
xi: logit bicycle i.agecat_old nonwhite nocar i.incomefifth i.urbancat5 sparse rf_dist_km rf_dist_kmsqrt rf_dist_kmsq ned_rf_avslope_perc interact interactsqrt /*
*/ if femagecat==`x' & home_gordet==`i' & flowtype <=2
}
forval i=10/11 { // SW and Wales = only larger cities
disp "femagecat `x' in region `i'"
xi: logit bicycle i.agecat_old nonwhite nocar i.incomefifth i.urbancat5_3only sparse rf_dist_km rf_dist_kmsqrt rf_dist_kmsq ned_rf_avslope_perc interact interactsqrt /*
*/ if femagecat==`x' & home_gordet==`i' & flowtype <=2
}
}
log close
*/
** FIT REGRESSION EQUATION - FLOW TYPE 1+2 - NEAR MARKET
tab agecat, gen(agecat_)
tab incomefifth, gen(incomefifth_)
tab urbancat5, gen(urbancat5_)
gen pred_basenmorig=.
replace pred_basenmorig = ((.292 * agecat_2) + (.524 * agecat_3) + (-.931 * nonwhite) + (.758 * nocar) + (-.050 * incomefifth_2) + (.035 * incomefifth_3) + (.031 * incomefifth_4) + (.094 * incomefifth_5) + (-.095 * urbancat5_3) + (-.226 * urbancat5_4) + (-.458 * urbancat5_5) + (.386 * sparse) + (-.652 * rf_dist_km) + (2.083 * rf_dist_kmsqrt) + (.009 * rf_dist_kmsq) + (-.224 * ned_rf_avslope_perc) + (.000 * interact) + (-.015 * interactsqrt) + -4.517) if flowtype<=2 & femagecat==1 & home_gordet==1
replace pred_basenmorig = ((.264 * agecat_2) + (.464 * agecat_3) + (-.918 * nonwhite) + (.788 * nocar) + (.073 * incomefifth_2) + (.048 * incomefifth_3) + (.103 * incomefifth_4) + (.038 * incomefifth_5) + (.281 * urbancat5_3) + (.384 * urbancat5_4) + (.225 * urbancat5_5) + (.140 * sparse) + (-.644 * rf_dist_km) + (2.089 * rf_dist_kmsqrt) + (.008 * rf_dist_kmsq) + (-.197 * ned_rf_avslope_perc) + (.030 * interact) + (-.141 * interactsqrt) + -4.553) if flowtype<=2 & femagecat==1 & home_gordet==2
replace pred_basenmorig = ((.249 * agecat_2) + (.450 * agecat_3) + (-.984 * nonwhite) + (.801 * nocar) + (.077 * incomefifth_2) + (.125 * incomefifth_3) + (.227 * incomefifth_4) + (.325 * incomefifth_5) + (.104 * urbancat5_2) + (.568 * urbancat5_3) + (.187 * urbancat5_4) + (.212 * urbancat5_5) + (.183 * sparse) + (-.751 * rf_dist_km) + (2.327 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.397 * ned_rf_avslope_perc) + (-.009 * interact) + (.078 * interactsqrt) + -4.705) if flowtype<=2 & femagecat==1 & home_gordet==3
replace pred_basenmorig = ((.161 * agecat_2) + (.347 * agecat_3) + (-.927 * nonwhite) + (.880 * nocar) + (-.045 * incomefifth_2) + (.050 * incomefifth_3) + (.097 * incomefifth_4) + (.156 * incomefifth_5) + (-.121 * urbancat5_2) + (-.246 * urbancat5_3) + (-.381 * urbancat5_4) + (-.494 * urbancat5_5) + (-.122 * sparse) + (-.708 * rf_dist_km) + (2.192 * rf_dist_kmsqrt) + (.010 * rf_dist_kmsq) + (-.143 * ned_rf_avslope_perc) + (.049 * interact) + (-.231 * interactsqrt) + -3.666) if flowtype<=2 & femagecat==1 & home_gordet==4
replace pred_basenmorig = ((.245 * agecat_2) + (.433 * agecat_3) + (-1.174 * nonwhite) + (.884 * nocar) + (.031 * incomefifth_2) + (.068 * incomefifth_3) + (.016 * incomefifth_4) + (.017 * incomefifth_5) + (.378 * urbancat5_3) + (.319 * urbancat5_4) + (.304 * urbancat5_5) + (-.171 * sparse) + (-.766 * rf_dist_km) + (2.367 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.101 * ned_rf_avslope_perc) + (.039 * interact) + (-.189 * interactsqrt) + -4.555) if flowtype<=2 & femagecat==1 & home_gordet==5
replace pred_basenmorig = ((.282 * agecat_2) + (.387 * agecat_3) + (-.567 * nonwhite) + (.934 * nocar) + (.082 * incomefifth_2) + (.255 * incomefifth_3) + (.247 * incomefifth_4) + (.500 * incomefifth_5) + (.665 * urbancat5_3) + (.400 * urbancat5_4) + (.296 * urbancat5_5) + (.181 * sparse) + (-.832 * rf_dist_km) + (2.564 * rf_dist_kmsqrt) + (.012 * rf_dist_kmsq) + (-.322 * ned_rf_avslope_perc) + (.017 * interact) + (-.122 * interactsqrt) + -4.772) if flowtype<=2 & femagecat==1 & home_gordet==6
replace pred_basenmorig = ((.468 * agecat_2) + (.517 * agecat_3) + (-.931 * nonwhite) + (.014 * nocar) + (.167 * incomefifth_2) + (.121 * incomefifth_3) + (-.107 * incomefifth_4) + (-.242 * incomefifth_5) + (-.612 * rf_dist_km) + (2.510 * rf_dist_kmsqrt) + (.005 * rf_dist_kmsq) + (.210 * ned_rf_avslope_perc) + (.081 * interact) + (-.473 * interactsqrt) + -4.580) if flowtype<=2 & femagecat==1 & home_gordet==7
replace pred_basenmorig = ((.481 * agecat_2) + (.676 * agecat_3) + (-1.192 * nonwhite) + (.391 * nocar) + (.154 * incomefifth_2) + (.205 * incomefifth_3) + (.302 * incomefifth_4) + (.468 * incomefifth_5) + (-.577 * urbancat5_3) + (-.198 * urbancat5_4) + (-.458 * urbancat5_5) + (-.376 * rf_dist_km) + (1.430 * rf_dist_kmsqrt) + (.003 * rf_dist_kmsq) + (-.286 * ned_rf_avslope_perc) + (.006 * interact) + (-.010 * interactsqrt) + -4.317) if flowtype<=2 & femagecat==1 & home_gordet==8
replace pred_basenmorig = ((.245 * agecat_2) + (.392 * agecat_3) + (-.644 * nonwhite) + (.753 * nocar) + (.007 * incomefifth_2) + (.107 * incomefifth_3) + (.145 * incomefifth_4) + (.170 * incomefifth_5) + (.283 * urbancat5_3) + (.146 * urbancat5_4) + (.196 * urbancat5_5) + (-.708 * rf_dist_km) + (2.202 * rf_dist_kmsqrt) + (.010 * rf_dist_kmsq) + (-.266 * ned_rf_avslope_perc) + (-.006 * interact) + (-.019 * interactsqrt) + -4.171) if flowtype<=2 & femagecat==1 & home_gordet==9
replace pred_basenmorig = ((.325 * agecat_2) + (.482 * agecat_3) + (-.435 * nonwhite) + (.564 * nocar) + (.112 * incomefifth_2) + (.101 * incomefifth_3) + (.236 * incomefifth_4) + (.305 * incomefifth_5) + (-.276 * urbancat5_4) + (-.401 * urbancat5_5) + (-.042 * sparse) + (-.767 * rf_dist_km) + (2.400 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.256 * ned_rf_avslope_perc) + (-.003 * interact) + (.053 * interactsqrt) + -4.064) if flowtype<=2 & femagecat==1 & home_gordet==10
replace pred_basenmorig = ((.400 * agecat_2) + (.626 * agecat_3) + (-.365 * nonwhite) + (.781 * nocar) + (-.026 * incomefifth_2) + (.027 * incomefifth_3) + (.178 * incomefifth_4) + (.164 * incomefifth_5) + (-.021 * urbancat5_4) + (-.261 * urbancat5_5) + (.386 * sparse) + (-.705 * rf_dist_km) + (2.385 * rf_dist_kmsqrt) + (.009 * rf_dist_kmsq) + (-.002 * ned_rf_avslope_perc) + (.044 * interact) + (-.267 * interactsqrt) + -4.945) if flowtype<=2 & femagecat==1 & home_gordet==11
replace pred_basenmorig = ((.522 * agecat_2) + (.557 * agecat_3) + (-.414 * nonwhite) + (.635 * nocar) + (.054 * incomefifth_2) + (.234 * incomefifth_3) + (.283 * incomefifth_4) + (.476 * incomefifth_5) + (-.213 * urbancat5_3) + (-.280 * urbancat5_4) + (-.489 * urbancat5_5) + (.371 * sparse) + (-.924 * rf_dist_km) + (2.633 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.911 * ned_rf_avslope_perc) + (-.062 * interact) + (.401 * interactsqrt) + -6.601) if flowtype<=2 & femagecat==11 & home_gordet==1
replace pred_basenmorig = ((.439 * agecat_2) + (.452 * agecat_3) + (-.584 * nonwhite) + (.803 * nocar) + (.144 * incomefifth_2) + (.074 * incomefifth_3) + (.191 * incomefifth_4) + (.142 * incomefifth_5) + (.270 * urbancat5_3) + (.251 * urbancat5_4) + (.305 * urbancat5_5) + (.828 * sparse) + (-.777 * rf_dist_km) + (2.409 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.494 * ned_rf_avslope_perc) + (.045 * interact) + (-.190 * interactsqrt) + -6.411) if flowtype<=2 & femagecat==11 & home_gordet==2
replace pred_basenmorig = ((.487 * agecat_2) + (.667 * agecat_3) + (-.756 * nonwhite) + (.725 * nocar) + (.090 * incomefifth_2) + (.164 * incomefifth_3) + (.368 * incomefifth_4) + (.537 * incomefifth_5) + (.312 * urbancat5_2) + (1.093 * urbancat5_3) + (.538 * urbancat5_4) + (.439 * urbancat5_5) + (.183 * sparse) + (-.975 * rf_dist_km) + (2.668 * rf_dist_kmsqrt) + (.018 * rf_dist_kmsq) + (-1.048 * ned_rf_avslope_perc) + (-.097 * interact) + (.505 * interactsqrt) + -6.414) if femagecat==11 & home_gordet==3
replace pred_basenmorig = ((.379 * agecat_2) + (.533 * agecat_3) + (-.972 * nonwhite) + (.725 * nocar) + (.056 * incomefifth_2) + (.151 * incomefifth_3) + (.199 * incomefifth_4) + (.124 * incomefifth_5) + (.040 * urbancat5_2) + (-.090 * urbancat5_3) + (-.266 * urbancat5_4) + (-.243 * urbancat5_5) + (-.199 * sparse) + (-.842 * rf_dist_km) + (2.268 * rf_dist_kmsqrt) + (.014 * rf_dist_kmsq) + (-.590 * ned_rf_avslope_perc) + (.033 * interact) + (-.142 * interactsqrt) + -4.988) if flowtype<=2 & femagecat==11 & home_gordet==4
replace pred_basenmorig = ((.466 * agecat_2) + (.496 * agecat_3) + (-.970 * nonwhite) + (.775 * nocar) + (.196 * incomefifth_2) + (.333 * incomefifth_3) + (.249 * incomefifth_4) + (.209 * incomefifth_5) + (.604 * urbancat5_3) + (.374 * urbancat5_4) + (.706 * urbancat5_5) + (.215 * sparse) + (-.860 * rf_dist_km) + (2.386 * rf_dist_kmsqrt) + (.014 * rf_dist_kmsq) + (-.418 * ned_rf_avslope_perc) + (.031 * interact) + (-.106 * interactsqrt) + -6.336) if flowtype<=2 & femagecat==11 & home_gordet==5
replace pred_basenmorig = ((.462 * agecat_2) + (.428 * agecat_3) + (-.501 * nonwhite) + (.927 * nocar) + (.303 * incomefifth_2) + (.551 * incomefifth_3) + (.528 * incomefifth_4) + (.881 * incomefifth_5) + (1.369 * urbancat5_3) + (.964 * urbancat5_4) + (.938 * urbancat5_5) + (.356 * sparse) + (-1.016 * rf_dist_km) + (2.950 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.422 * ned_rf_avslope_perc) + (.053 * interact) + (-.360 * interactsqrt) + -6.755) if flowtype<=2 & femagecat==11 & home_gordet==6
replace pred_basenmorig = ((.632 * agecat_2) + (.519 * agecat_3) + (-1.036 * nonwhite) + (.011 * nocar) + (.146 * incomefifth_2) + (.085 * incomefifth_3) + (-.070 * incomefifth_4) + (-.220 * incomefifth_5) + (-.891 * rf_dist_km) + (3.395 * rf_dist_kmsqrt) + (.010 * rf_dist_kmsq) + (.148 * ned_rf_avslope_perc) + (.086 * interact) + (-.503 * interactsqrt) + -5.958) if flowtype<=2 & femagecat==11 & home_gordet==7
replace pred_basenmorig = ((.803 * agecat_2) + (.769 * agecat_3) + (-1.291 * nonwhite) + (.552 * nocar) + (.257 * incomefifth_2) + (.328 * incomefifth_3) + (.392 * incomefifth_4) + (.689 * incomefifth_5) + (-.724 * urbancat5_3) + (-.567 * urbancat5_4) + (-.306 * urbancat5_5) + (-.486 * rf_dist_km) + (1.857 * rf_dist_kmsqrt) + (.003 * rf_dist_kmsq) + (-.321 * ned_rf_avslope_perc) + (.013 * interact) + (-.068 * interactsqrt) + -6.149) if flowtype<=2 & femagecat==11 & home_gordet==8
replace pred_basenmorig = ((.466 * agecat_2) + (.444 * agecat_3) + (-.611 * nonwhite) + (.815 * nocar) + (.159 * incomefifth_2) + (.380 * incomefifth_3) + (.417 * incomefifth_4) + (.433 * incomefifth_5) + (.557 * urbancat5_3) + (.123 * urbancat5_4) + (.245 * urbancat5_5) + (-.961 * rf_dist_km) + (2.766 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.333 * ned_rf_avslope_perc) + (.005 * interact) + (-.100 * interactsqrt) + -6.076) if flowtype<=2 & femagecat==11 & home_gordet==9
replace pred_basenmorig = ((.545 * agecat_2) + (.490 * agecat_3) + (-.338 * nonwhite) + (.555 * nocar) + (.141 * incomefifth_2) + (-.012 * incomefifth_3) + (.211 * incomefifth_4) + (.271 * incomefifth_5) + (-.504 * urbancat5_4) + (-.390 * urbancat5_5) + (.207 * sparse) + (-1.044 * rf_dist_km) + (2.975 * rf_dist_kmsqrt) + (.018 * rf_dist_kmsq) + (-.439 * ned_rf_avslope_perc) + (-.023 * interact) + (.164 * interactsqrt) + -5.516) if flowtype<=2 & femagecat==11 & home_gordet==10
replace pred_basenmorig = ((.554 * agecat_2) + (.365 * agecat_3) + (-.176 * nonwhite) + (.732 * nocar) + (.041 * incomefifth_2) + (.054 * incomefifth_3) + (.730 * incomefifth_4) + (.499 * incomefifth_5) + (-.266 * urbancat5_4) + (-.324 * urbancat5_5) + (.606 * sparse) + (-1.151 * rf_dist_km) + (3.694 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.236 * ned_rf_avslope_perc) + (.088 * interact) + (-.378 * interactsqrt) + -7.285) if flowtype<=2 & femagecat==11 & home_gordet==11
replace pred_basenmorig = ((-.625 * agecat_5) + (-.336 * agecat_6) + (-1.234 * nonwhite) + (1.084 * nocar) + (.027 * incomefifth_2) + (.074 * incomefifth_3) + (.084 * incomefifth_4) + (.215 * incomefifth_5) + (-.059 * urbancat5_3) + (-.131 * urbancat5_4) + (-.271 * urbancat5_5) + (.406 * sparse) + (-.595 * rf_dist_km) + (1.851 * rf_dist_kmsqrt) + (.008 * rf_dist_kmsq) + (-.310 * ned_rf_avslope_perc) + (.024 * interact) + (-.079 * interactsqrt) + -4.273) if flowtype<=2 & femagecat==2 & home_gordet==1
replace pred_basenmorig = ((-.609 * agecat_5) + (-.376 * agecat_6) + (-1.071 * nonwhite) + (1.085 * nocar) + (.122 * incomefifth_2) + (.113 * incomefifth_3) + (.165 * incomefifth_4) + (.178 * incomefifth_5) + (.397 * urbancat5_3) + (.557 * urbancat5_4) + (.373 * urbancat5_5) + (.257 * sparse) + (-.593 * rf_dist_km) + (1.739 * rf_dist_kmsqrt) + (.008 * rf_dist_kmsq) + (-.301 * ned_rf_avslope_perc) + (.022 * interact) + (-.080 * interactsqrt) + -4.164) if flowtype<=2 & femagecat==2 & home_gordet==2
replace pred_basenmorig = ((-.600 * agecat_5) + (-.445 * agecat_6) + (-.871 * nonwhite) + (1.099 * nocar) + (.031 * incomefifth_2) + (.127 * incomefifth_3) + (.187 * incomefifth_4) + (.407 * incomefifth_5) + (.234 * urbancat5_2) + (.846 * urbancat5_3) + (.490 * urbancat5_4) + (.262 * urbancat5_5) + (.129 * sparse) + (-.752 * rf_dist_km) + (2.052 * rf_dist_kmsqrt) + (.013 * rf_dist_kmsq) + (-.617 * ned_rf_avslope_perc) + (-.034 * interact) + (.204 * interactsqrt) + -4.321) if flowtype<=2 & femagecat==2 & home_gordet==3
replace pred_basenmorig = ((-.592 * agecat_5) + (-.176 * agecat_6) + (-1.063 * nonwhite) + (1.185 * nocar) + (.110 * incomefifth_2) + (.127 * incomefifth_3) + (.239 * incomefifth_4) + (.277 * incomefifth_5) + (.215 * urbancat5_2) + (.133 * urbancat5_3) + (-.108 * urbancat5_4) + (-.314 * urbancat5_5) + (.196 * sparse) + (-.628 * rf_dist_km) + (1.645 * rf_dist_kmsqrt) + (.009 * rf_dist_kmsq) + (-.355 * ned_rf_avslope_perc) + (.030 * interact) + (-.106 * interactsqrt) + -3.380) if flowtype<=2 & femagecat==2 & home_gordet==4
replace pred_basenmorig = ((-.627 * agecat_5) + (-.374 * agecat_6) + (-1.212 * nonwhite) + (1.210 * nocar) + (.062 * incomefifth_2) + (.144 * incomefifth_3) + (.131 * incomefifth_4) + (.205 * incomefifth_5) + (.486 * urbancat5_3) + (.453 * urbancat5_4) + (.257 * urbancat5_5) + (.156 * sparse) + (-.763 * rf_dist_km) + (2.101 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.254 * ned_rf_avslope_perc) + (.038 * interact) + (-.132 * interactsqrt) + -4.181) if flowtype<=2 & femagecat==2 & home_gordet==5
replace pred_basenmorig = ((-.487 * agecat_5) + (-.482 * agecat_6) + (-.612 * nonwhite) + (1.291 * nocar) + (.147 * incomefifth_2) + (.224 * incomefifth_3) + (.226 * incomefifth_4) + (.467 * incomefifth_5) + (.684 * urbancat5_3) + (.440 * urbancat5_4) + (.315 * urbancat5_5) + (.488 * sparse) + (-.766 * rf_dist_km) + (2.115 * rf_dist_kmsqrt) + (.012 * rf_dist_kmsq) + (-.365 * ned_rf_avslope_perc) + (-.009 * interact) + (-.040 * interactsqrt) + -4.151) if flowtype<=2 & femagecat==2 & home_gordet==6
replace pred_basenmorig = ((-.751 * agecat_5) + (-.457 * agecat_6) + (-1.089 * nonwhite) + (.126 * nocar) + (.256 * incomefifth_2) + (.324 * incomefifth_3) + (.242 * incomefifth_4) + (.053 * incomefifth_5) + (-.561 * rf_dist_km) + (2.228 * rf_dist_kmsqrt) + (.005 * rf_dist_kmsq) + (.133 * ned_rf_avslope_perc) + (.023 * interact) + (-.296 * interactsqrt) + -4.342) if flowtype<=2 & femagecat==2 & home_gordet==7
replace pred_basenmorig = ((-.776 * agecat_5) + (-.393 * agecat_6) + (-1.312 * nonwhite) + (.682 * nocar) + (.071 * incomefifth_2) + (.247 * incomefifth_3) + (.329 * incomefifth_4) + (.488 * incomefifth_5) + (-.456 * urbancat5_3) + (-.265 * urbancat5_4) + (.031 * urbancat5_5) + (-.449 * rf_dist_km) + (1.421 * rf_dist_kmsqrt) + (.005 * rf_dist_kmsq) + (-.402 * ned_rf_avslope_perc) + (.000 * interact) + (.014 * interactsqrt) + -3.852) if flowtype<=2 & femagecat==2 & home_gordet==8
replace pred_basenmorig = ((-.549 * agecat_5) + (-.402 * agecat_6) + (-.626 * nonwhite) + (1.083 * nocar) + (.042 * incomefifth_2) + (.157 * incomefifth_3) + (.234 * incomefifth_4) + (.213 * incomefifth_5) + (.308 * urbancat5_3) + (.062 * urbancat5_4) + (-.069 * urbancat5_5) + (-.643 * rf_dist_km) + (1.792 * rf_dist_kmsqrt) + (.010 * rf_dist_kmsq) + (-.327 * ned_rf_avslope_perc) + (.009 * interact) + (-.054 * interactsqrt) + -3.661) if flowtype<=2 & femagecat==2 & home_gordet==9
replace pred_basenmorig = ((-.708 * agecat_5) + (-.481 * agecat_6) + (-.456 * nonwhite) + (.865 * nocar) + (.083 * incomefifth_2) + (.168 * incomefifth_3) + (.223 * incomefifth_4) + (.293 * incomefifth_5) + (-.199 * urbancat5_4) + (-.386 * urbancat5_5) + (-.215 * sparse) + (-.713 * rf_dist_km) + (1.950 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.273 * ned_rf_avslope_perc) + (.008 * interact) + (.004 * interactsqrt) + -3.344) if flowtype<=2 & femagecat==2 & home_gordet==10
replace pred_basenmorig = ((-.612 * agecat_5) + (-.249 * agecat_6) + (-.360 * nonwhite) + (1.059 * nocar) + (-.105 * incomefifth_2) + (.013 * incomefifth_3) + (.189 * incomefifth_4) + (.258 * incomefifth_5) + (-.003 * urbancat5_4) + (-.296 * urbancat5_5) + (.351 * sparse) + (-.537 * rf_dist_km) + (1.520 * rf_dist_kmsqrt) + (.008 * rf_dist_kmsq) + (-.278 * ned_rf_avslope_perc) + (.014 * interact) + (-.064 * interactsqrt) + -3.878) if flowtype<=2 & femagecat==2 & home_gordet==11
replace pred_basenmorig = ((-.279 * agecat_5) + (.510 * agecat_6) + (-1.170 * nonwhite) + (.496 * nocar) + (.071 * incomefifth_2) + (.004 * incomefifth_3) + (.044 * incomefifth_4) + (.220 * incomefifth_5) + (.008 * urbancat5_3) + (-.008 * urbancat5_4) + (-.023 * urbancat5_5) + (.615 * sparse) + (-.882 * rf_dist_km) + (2.497 * rf_dist_kmsqrt) + (.015 * rf_dist_kmsq) + (-.781 * ned_rf_avslope_perc) + (.007 * interact) + (.023 * interactsqrt) + -6.004) if flowtype<=2 & femagecat==12 & home_gordet==1
replace pred_basenmorig = ((.014 * agecat_5) + (.496 * agecat_6) + (-.556 * nonwhite) + (.717 * nocar) + (.098 * incomefifth_2) + (.077 * incomefifth_3) + (.192 * incomefifth_4) + (.214 * incomefifth_5) + (.755 * urbancat5_3) + (.695 * urbancat5_4) + (1.119 * urbancat5_5) + (.511 * sparse) + (-.602 * rf_dist_km) + (1.442 * rf_dist_kmsqrt) + (.011 * rf_dist_kmsq) + (-.792 * ned_rf_avslope_perc) + (-.006 * interact) + (.056 * interactsqrt) + -5.288) if flowtype<=2 & femagecat==12 & home_gordet==2
replace pred_basenmorig = ((-.228 * agecat_5) + (.379 * agecat_6) + (-.914 * nonwhite) + (.550 * nocar) + (.052 * incomefifth_2) + (.000 * incomefifth_3) + (.115 * incomefifth_4) + (.215 * incomefifth_5) + (.614 * urbancat5_2) + (1.550 * urbancat5_3) + (1.221 * urbancat5_4) + (.955 * urbancat5_5) + (.194 * sparse) + (-.857 * rf_dist_km) + (1.969 * rf_dist_kmsqrt) + (.018 * rf_dist_kmsq) + (-1.291 * ned_rf_avslope_perc) + (-.096 * interact) + (.543 * interactsqrt) + -5.191) if flowtype<=2 & femagecat==12 & home_gordet==3
replace pred_basenmorig = ((-.149 * agecat_5) + (.421 * agecat_6) + (-1.332 * nonwhite) + (.668 * nocar) + (.098 * incomefifth_2) + (.148 * incomefifth_3) + (.242 * incomefifth_4) + (.181 * incomefifth_5) + (1.082 * urbancat5_2) + (1.249 * urbancat5_3) + (1.271 * urbancat5_4) + (1.109 * urbancat5_5) + (.146 * sparse) + (-.778 * rf_dist_km) + (1.723 * rf_dist_kmsqrt) + (.015 * rf_dist_kmsq) + (-.910 * ned_rf_avslope_perc) + (.011 * interact) + (.000 * interactsqrt) + -5.054) if flowtype<=2 & femagecat==12 & home_gordet==4
replace pred_basenmorig = ((-.036 * agecat_5) + (.768 * agecat_6) + (-.963 * nonwhite) + (.576 * nocar) + (.092 * incomefifth_2) + (.266 * incomefifth_3) + (.232 * incomefifth_4) + (.282 * incomefifth_5) + (.932 * urbancat5_3) + (1.074 * urbancat5_4) + (1.075 * urbancat5_5) + (.198 * sparse) + (-.765 * rf_dist_km) + (1.666 * rf_dist_kmsqrt) + (.014 * rf_dist_kmsq) + (-.680 * ned_rf_avslope_perc) + (.032 * interact) + (-.051 * interactsqrt) + -5.112) if flowtype<=2 & femagecat==12 & home_gordet==5
replace pred_basenmorig = ((-.136 * agecat_5) + (.291 * agecat_6) + (-.704 * nonwhite) + (.736 * nocar) + (.144 * incomefifth_2) + (.316 * incomefifth_3) + (.211 * incomefifth_4) + (.465 * incomefifth_5) + (1.355 * urbancat5_3) + (1.356 * urbancat5_4) + (1.228 * urbancat5_5) + (.698 * sparse) + (-.809 * rf_dist_km) + (1.852 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.579 * ned_rf_avslope_perc) + (.021 * interact) + (-.185 * interactsqrt) + -4.928) if flowtype<=2 & femagecat==12 & home_gordet==6
replace pred_basenmorig = ((-.629 * agecat_5) + (-.296 * agecat_6) + (-1.178 * nonwhite) + (.080 * nocar) + (.444 * incomefifth_2) + (.571 * incomefifth_3) + (.534 * incomefifth_4) + (.379 * incomefifth_5) + (-.746 * rf_dist_km) + (2.764 * rf_dist_kmsqrt) + (.009 * rf_dist_kmsq) + (-.487 * ned_rf_avslope_perc) + (-.150 * interact) + (.377 * interactsqrt) + -5.662) if flowtype<=2 & femagecat==12 & home_gordet==7
replace pred_basenmorig = ((-.638 * agecat_5) + (-.045 * agecat_6) + (-1.433 * nonwhite) + (.642 * nocar) + (.381 * incomefifth_2) + (.439 * incomefifth_3) + (.613 * incomefifth_4) + (1.062 * incomefifth_5) + (-.293 * urbancat5_3) + (-.028 * urbancat5_4) + (-1.017 * urbancat5_5) + (-.626 * rf_dist_km) + (1.778 * rf_dist_kmsqrt) + (.009 * rf_dist_kmsq) + (-.615 * ned_rf_avslope_perc) + (.024 * interact) + (-.071 * interactsqrt) + -5.300) if flowtype<=2 & femagecat==12 & home_gordet==8
replace pred_basenmorig = ((-.130 * agecat_5) + (.518 * agecat_6) + (-.659 * nonwhite) + (.770 * nocar) + (.125 * incomefifth_2) + (.254 * incomefifth_3) + (.330 * incomefifth_4) + (.352 * incomefifth_5) + (.500 * urbancat5_3) + (.352 * urbancat5_4) + (.421 * urbancat5_5) + (-.759 * rf_dist_km) + (1.733 * rf_dist_kmsqrt) + (.014 * rf_dist_kmsq) + (-.610 * ned_rf_avslope_perc) + (.013 * interact) + (-.048 * interactsqrt) + -4.441) if flowtype<=2 & femagecat==12 & home_gordet==9
replace pred_basenmorig = ((-.253 * agecat_5) + (.525 * agecat_6) + (-.490 * nonwhite) + (.431 * nocar) + (.167 * incomefifth_2) + (.050 * incomefifth_3) + (.248 * incomefifth_4) + (.342 * incomefifth_5) + (-.116 * urbancat5_4) + (-.070 * urbancat5_5) + (-.159 * sparse) + (-.815 * rf_dist_km) + (1.746 * rf_dist_kmsqrt) + (.016 * rf_dist_kmsq) + (-.768 * ned_rf_avslope_perc) + (-.034 * interact) + (.271 * interactsqrt) + -3.705) if flowtype<=2 & femagecat==12 & home_gordet==10
replace pred_basenmorig = ((-.478 * agecat_5) + (.705 * agecat_6) + (-.364 * nonwhite) + (.711 * nocar) + (-.063 * incomefifth_2) + (-.119 * incomefifth_3) + (.318 * incomefifth_4) + (.323 * incomefifth_5) + (.261 * urbancat5_4) + (-.270 * urbancat5_5) + (.736 * sparse) + (-.780 * rf_dist_km) + (2.043 * rf_dist_kmsqrt) + (.012 * rf_dist_kmsq) + (-.717 * ned_rf_avslope_perc) + (.030 * interact) + (.002 * interactsqrt) + -5.426) if flowtype<=2 & femagecat==12 & home_gordet==11
**CONVERT LOG-ODDS TO PROBABILITIES
foreach x in basegt dutch ebike basenmorig {
replace pred_`x'=exp(pred_`x')/(1+exp(pred_`x'))
}
/** UPDATED DUTCH AND EBIKES - FIG 2 + FIG 3 IN USER MANUAL C
gen rf_dist_kmround=(floor(rf_dist_km))
gen rf_avslope_percround=floor(rf_avslope_perc*10)/10
recode rf_avslope_percround 4/max=4
table rf_dist_kmround if flowtype<=2 , c(mean bicycle mean pred_basegt)
table rf_avslope_percround if flowtype<=2 , c(mean bicycle mean pred_basegt)
table rf_dist_kmround if flowtype<=2 , c(mean bicycle mean pred_dutch mean pred_ebike)
table rf_avslope_percround if flowtype<=2 , c(mean bicycle mean pred_dutch mean pred_ebike)
*/
** INPUT PARAMETERS FLOW 3 (NO FIXED PLACE)
foreach x in pred_basegt pred_basenmorig bdutch bebike { // WEIGHTED AVERAGE OF COMMUTERS IN FLOWTYPE 1 & "
gen `x'temp=`x'
replace `x'temp=. if flowtype!=1 & flowtype!=2 // WEIGHTED AVERAGE OF COMMUTERS IN FLOWTYPE 1 & "
gen alltemp=1
replace alltemp=. if `x'temp==.
bysort home_lsoa: egen nummean`x'=sum(`x'temp*alltemp)
bysort home_lsoa: egen denmean`x'=sum(alltemp)
gen mean`x'=nummean`x'/denmean`x'
drop `x'temp alltemp nummean`x' denmean`x'
}
gen meanpred_basegtsq=meanpred_basegt^2
gen meanpred_basegtsqrt=meanpred_basegt^0.5
gen meanpred_basenmorigsq=meanpred_basenmorig^2
gen meanpred_basenmorigsqrt=meanpred_basenmorig^0.5
** FIT REGRESSION EQUATION - FLOW 3 - GOV TARGET, DUTCH, EBIKE
*xi: logit bicycle meanpred_basegtsq meanpred_basegtsqrt if flowtype==3
gen pred2_basegt= -6.530 + (132.2 * meanpred_basegtsq) + (11.47 * meanpred_basegtsqrt)
gen pred2_dutch= pred2_basegt + meanbdutch
gen pred2_ebike= pred2_dutch + meanbebike
/* IDENTIFY REGRESSION EQUATION - FLOW TYPE 1+2 - NEAR MARKET. Paste to 'pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\1logfiles\NearMarket_flow3.xls'
log using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\1logfiles\NearMarket_flow3.smcl", replace
forval i=1/11 {
disp "region `i'"
xi: logit bicycle meanpred_basenmorigsq meanpred_basenmorigsqrt if flowtype==3 & home_gordet==`i'
}
log close
*/
** FIT REGRESSION EQUATION - FLOW 3 - NEAR MARKET
gen pred2_basenmorig=.
replace pred2_basenmorig = ((351.2 * meanpred_basenmorigsq) + (18.33 * meanpred_basenmorigsqrt) + -7.35) if flowtype==3 & home_gordet==1
replace pred2_basenmorig = ((-509.1 * meanpred_basenmorigsq) + (26.06 * meanpred_basenmorigsqrt) + -8.13) if flowtype==3 & home_gordet==2
replace pred2_basenmorig = ((7.3 * meanpred_basenmorigsq) + (10.74 * meanpred_basenmorigsqrt) + -6.32) if flowtype==3 & home_gordet==3
replace pred2_basenmorig = ((-178.9 * meanpred_basenmorigsq) + (19.14 * meanpred_basenmorigsqrt) + -7.61) if flowtype==3 & home_gordet==4
replace pred2_basenmorig = ((182.5 * meanpred_basenmorigsq) + (8.37 * meanpred_basenmorigsqrt) + -5.98) if flowtype==3 & home_gordet==5
replace pred2_basenmorig = ((66.5 * meanpred_basenmorigsq) + (5.79 * meanpred_basenmorigsqrt) + -5.71) if flowtype==3 & home_gordet==6
replace pred2_basenmorig = ((-178.4 * meanpred_basenmorigsq) + (28.68 * meanpred_basenmorigsqrt) + -9.64) if flowtype==3 & home_gordet==7
replace pred2_basenmorig = ((83.5 * meanpred_basenmorigsq) + (19.06 * meanpred_basenmorigsqrt) + -7.45) if flowtype==3 & home_gordet==8
replace pred2_basenmorig = ((118.2 * meanpred_basenmorigsq) + (7.65 * meanpred_basenmorigsqrt) + -5.98) if flowtype==3 & home_gordet==9
replace pred2_basenmorig = ((-93.0 * meanpred_basenmorigsq) + (15.50 * meanpred_basenmorigsqrt) + -7.14) if flowtype==3 & home_gordet==10
replace pred2_basenmorig = ((114.6 * meanpred_basenmorigsq) + (14.82 * meanpred_basenmorigsqrt) + -6.88) if flowtype==3 & home_gordet==11
** COMBINE FLOW 3+4 WITH 1+2
foreach x in basegt dutch ebike basenmorig {
replace pred2_`x'=exp(pred2_`x')/(1+exp(pred2_`x'))
replace pred_`x'=pred2_`x' if flowtype==3
drop pred2_`x'
replace pred_`x'=0 if flowtype==4
}
** SCALE BY REGION TO BE EQIVALENT TO GOV TARGET. Paste to 'pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\1logfiles\NearMarket_gtscale.xls'
table home_gordet , c(mean bicycle mean pred_basegt mean pred_basenmorig)
gen gtscaling=. // pred_basegt divided by pred_basenmorig
recode gtscaling .=1.731 if home_gordet==1
recode gtscaling .=1.550 if home_gordet==2
recode gtscaling .=1.032 if home_gordet==3
recode gtscaling .=1.040 if home_gordet==4
recode gtscaling .=1.382 if home_gordet==5
recode gtscaling .=0.824 if home_gordet==6
recode gtscaling .=0.592 if home_gordet==7
recode gtscaling .=1.406 if home_gordet==8
recode gtscaling .=0.906 if home_gordet==9
recode gtscaling .=0.665 if home_gordet==10
recode gtscaling .=1.422 if home_gordet==11
gen pred_basenm=pred_basenmorig*gtscaling
*table home_gordet , c(mean bicycle mean pred_basegt mean pred_basenmorig mean pred_basenm)
** SAVE
keep home_lsoa- incomedecile bicycle pred_basegt pred_basenmorig pred_basenm pred_dutch pred_ebike
compress
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\CommuteSPTemp2.dta", replace
*****************
** PART 3B: APPLY SCENARIOS TO DATA: SLC AND SIC
*****************
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\CommuteSPTemp2.dta", clear
* NO CYCLISTS
gen nocyclists_slc=0
gen nocyclists_sic=nocyclists_slc-bicycle
* GOV TARGET, NEAR MARKET - CALC SIC THEN ADD TO BASELINE FOR SLC
foreach x in basegt basenm {
bysort home_lsoa work_lsoa: egen odcyclist_all`x'=sum(pred_`x') // NO. PREDICTED NEW CYCLISTS IN FLOW IN SCENARIO IN TOTAL
bysort home_lsoa work_lsoa: egen odcyclist_noncyclist`x'=sum((1-bicycle)*pred_`x') // NO. PREDICTED NEW CYCLISTS IN FLOW IN SCENARIO IF RESTRICTED TO CURRENT NONCYCLISTS
gen `x'_sic=pred_`x'*(odcyclist_all`x'/odcyclist_noncyclist`x') // SCALING FACTOR - INCREASE THE SIC AMONG NON-CYCLISTS SO THAT CAN BE ZERO IN CURRENT CYCLISTS
replace `x'_sic=0 if bicycle==1 // NO SIC IN CYCLISTS
recode `x'_sic 1/max=1 // A SMALL NUMBER OF FLOWS WHICH ARE ALMOST ALL CYCLISTS OTHERWISE HAVE THIS >1
gen `x'_slc=`x'_sic+bicycle
drop odcyclist_all`x' odcyclist_noncyclist`x'
}
rename basegt* govtarget*
rename basenm* govnearmkt*
* GENDER EQ
bysort home_lsoa work_lsoa: egen bicycle_all=sum(bicycle)
bysort home_lsoa work_lsoa: egen bicycle_male=sum((1-female)*bicycle)
bysort home_lsoa work_lsoa: egen bicycle_female=sum((female)*bicycle)
bysort home_lsoa work_lsoa: egen female_all=sum((female))
bysort home_lsoa work_lsoa: egen male_all=sum((1-female))
gen gendereq_totalslc=(bicycle_male*(1 + (female_all/male_all))) // SLC IN FLOW AS WHOLE
gen gendereq_sic=(gendereq_totalslc-bicycle_all)/(female_all-bicycle_female) // SIC PER FEMALE NON-CYCLIST
replace gendereq_sic = 0 if female_all==bicycle_female // NO CHANGE IF ALL WOMEN ALREADY CYCLE
replace gendereq_sic = 0 if gendereq_totalslc<=bicycle_all // NO CHANGE IF SLC<=BASELINE
replace gendereq_sic = 0 if bicycle_male==0 // NO CHANGE IF NO MALE CYCLISTS IN FLOW
replace gendereq_sic = 0 if female==0 | bicycle==1 // NO CHANGE AMONG MALES OR AMONG FEMALE CYCLISTS
gen gendereq_slc=gendereq_sic+bicycle
drop bicycle_all- male_all gendereq_totalslc
* DUTCH + EBIKE - CALC SLC, THEN MINUS BASELINE TO GET SIC
bysort home_lsoa work_lsoa: gen odall=_N // NO. COMMUTERS IN FLOW
bysort home_lsoa work_lsoa: egen odbicycle=sum(bicycle) // NO. CYCLISTS IN FLOW
foreach x in dutch ebike {
bysort home_lsoa work_lsoa: egen odcyclist_all`x'=sum(pred_`x') // NO. PREDICTED NEW CYCLISTS IN FLOW IN SCENARIO IN TOTAL
gen `x'_slc=(odcyclist_all`x'-odbicycle)/(odall-odbicycle) // INCREASE IN NON-CYCLISTS TO GIVE RIGHT TOTAL
replace `x'_slc=1 if bicycle==1 // SLC = BASELINE CYCLING IN CYCLISTS
gen `x'_sic=`x'_slc-bicycle
drop odcyclist_all`x'
}
drop odall odbicycle
* NO INCREASE IN FLOW TYPE 4 (TOO LONG/OUTSIDE EW)
foreach x in govtarget govnearmkt gendereq dutch ebike {
replace `x'_slc=bicycle if flowtype==4
replace `x'_sic=0 if flowtype==4
}
*****************
** PART 3C: MODE SHIFT
*****************
recode commute_mainmode9 1=0 2=1 3/max=0,gen(foot)
recode commute_mainmode9 1/2=0 3=1 4/max=0,gen(car_driver)
recode commute_mainmode9 1/3=0 4=1 5/max=0,gen(car_passenger)
recode commute_mainmode9 1/4=0 5=1 6/max=0,gen(motorbike)
recode commute_mainmode9 1/5=0 6/7=1 8/max=0,gen(public_transport)
bysort home_lsoa work_lsoa: gen od_all=_N
foreach x in bicycle foot car_driver car_passenger motorbike public_transport {
bysort home_lsoa work_lsoa: egen od_`x'=sum(`x')
}
* NO CYCLISTS
gen nocyclists_slw=foot
replace nocyclists_slw=od_foot/(od_all-od_bicycle) if bicycle==1 // Cyclists get walk mode share equal to flow mode share among non-cyclists
replace nocyclists_slw=0.31 if od_bicycle==od_all // Flows with pure bicycles at baseline - make walking 31% of new flows [from Lovelace 2017, based on MSOA mode pairs with 50%-99% cycling]
gen nocyclists_siw=nocyclists_slw-foot
gen nocyclists_sld=car_driver
replace nocyclists_sld=od_car_driver/(od_all-od_bicycle) if bicycle==1 // Cyclists get walk mode share equal to flow mode share among non-cyclists
replace nocyclists_sld=0.35 if od_bicycle==od_all // Flows with pure bicycles at baseline - make driving 35% of new flows [from Lovelace 2017, based on MSOA mode pairs with 50%-99% cycling]
gen nocyclists_sid=nocyclists_sld-car_driver
gen nocyclists_slp=car_passenger
replace nocyclists_slp=od_car_passenger/(od_all-od_bicycle) if bicycle==1 // Cyclists get walk mode share equal to flow mode share among non-cyclists
replace nocyclists_slp=0.04 if od_bicycle==od_all // Flows with pure bicycles at baseline - make car pass 4% of new flows [same approach as Lovelace 2017, based on MSOA mode pairs with 50%-99% cycling]
gen nocyclists_sip=nocyclists_slp-car_passenger
gen nocyclists_slm=motorbike
replace nocyclists_slm=od_motorbike/(od_all-od_bicycle) if bicycle==1 // Cyclists get walk mode share equal to flow mode share among non-cyclists
replace nocyclists_slm=0.02 if od_bicycle==od_all // Flows with pure bicycles at baseline - make mbike 2% of new flows [same approach as Lovelace 2017, based on MSOA mode pairs with 50%-99% cycling]
gen nocyclists_sim=nocyclists_slm-motorbike
gen nocyclists_slpt=public_transport
replace nocyclists_slpt=od_public_transport/(od_all-od_bicycle) if bicycle==1 // Cyclists get walk mode share equal to flow mode share among non-cyclists
replace nocyclists_slpt=0.28 if od_bicycle==od_all // Flows with pure bicycles at baseline - make pub transport 28% of new flows [same approach as Lovelace 2017, based on MSOA mode pairs with 50%-99% cycling]
gen nocyclists_sipt=nocyclists_slpt-public_transport
* BY SCENARIO
foreach x in govtarget govnearmkt gendereq dutch ebike {
gen `x'_siw=foot*`x'_sic*-1
gen `x'_slw=foot+`x'_siw
gen `x'_sid=car_driver*`x'_sic*-1
gen `x'_sld=car_driver+`x'_sid
gen `x'_sip=car_passenger*`x'_sic*-1
gen `x'_slp=car_passenger+`x'_sip
gen `x'_sim=motorbike*`x'_sic*-1
gen `x'_slm=motorbike+`x'_sim
gen `x'_sipt=public_transport*`x'_sic*-1
gen `x'_slpt=public_transport+`x'_sipt
}
* NO INCREASE IN DRIVING FLOW TYPE 4 (TOO LONG/OUTSIDE EW) - SO GET RIGHT FOR INCREASE
foreach x in govtarget govnearmkt gendereq dutch ebike {
replace `x'_sld=car_driver if flowtype==4
replace `x'_sid=0 if flowtype==4
}
* ORDER + DROP EXCESS VARIABLES
drop pred_basegt pred_dutch pred_ebike pred_basenmorig pred_basenm
drop bicycle foot- od_public_transport
foreach x in ebike dutch gendereq govnearmkt govtarget nocyclists {
order `x'_slc `x'_sic `x'_slw `x'_siw `x'_sld `x'_sid `x'_slp `x'_sip `x'_slm `x'_sim `x'_slpt `x'_sipt, after(incomedecile)
}
*****************
** STEP 4: DO TAG AND CARBON
*****************
** MERGE IN GRADIENT + SICKNESS ABSENCE + SALARY
gen gradient=0.25* (round(cyc_avslope_perc*4))
recode gradient 7/max=7
merge m:1 gradient using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_mmetspeed_hilliness.dta", nogen
recode agecat 6=5, gen(agecat5)
merge m:1 female agecat5 home_gor using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_sickness_hours_year.dta", nogen
merge m:1 home_gor using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\0temp\EngWales_salary_hourly.dta", nogen
merge m:1 female agecat home_gor using "pct-inputs\01_raw\04_other_data\GBD_YLLdeaths\GBD_YLLperDeath.dta" ,nogen
** INPUT PARAMETERS TAG + CARBON [APPENDIX TABLE]
gen cyclecommute_tripspertypicalweek = .
replace cyclecommute_tripspertypicalweek = 7.24 if female==0 & agecat<=3
replace cyclecommute_tripspertypicalweek = 7.32 if female==0 & agecat>=4
replace cyclecommute_tripspertypicalweek = 6.31 if female==1 & agecat<=3
replace cyclecommute_tripspertypicalweek = 7.23 if female==1 & agecat>=4
gen cyclecommute_tripsperweek = .
replace cyclecommute_tripsperweek = 5.46 if female==0 & agecat<=3
replace cyclecommute_tripsperweek = 5.23 if female==0 & agecat>=4
replace cyclecommute_tripsperweek = 4.13 if female==1 & agecat<=3
replace cyclecommute_tripsperweek = 4.88 if female==1 & agecat>=4
recode cyc_dist_km min/4.9999999=.07 5/9.9999999=.13 10/19.999999=.23 20/max=.23, gen(percentebike_dutch)
recode cyc_dist_km min/4.9999999=.71 5/9.9999999=.91 10/19.999999=.93 20/max=1, gen(percentebike_ebike)
gen rr_ref_pa=0.9
gen mmet_ref=8.75
gen vsly = 57965
gen rr_ref_sick=0.75
gen co2kg_km=0.182
** ANNUAL DURATION OF CYCLING PER CYCLIST
foreach x in cycle ebike {
gen hrs_`x' = ((cyc_dist_km*cyclecommute_tripsperweek)/speed_`x') * 52.2 // CYCLING PER YEAR IN HOURS AMONG NEW CYCLISTS
}
foreach x in nocyclists govtarget govnearmkt gendereq {
gen `x'_sicyclehours = `x'_sic * hrs_cycle
}
foreach x in dutch ebike {
gen `x'_sicyclehours=`x'_sic *(((1-percentebike_`x')*hrs_cycle)+(percentebike_`x'*hrs_ebike))
}
gen base_slcyclehours = -1 * nocyclists_sicyclehours
order base_slcyclehours, before(govtarget_sicyclehours)
** WEEKLY mMETS OF CYCLING/WALKING
foreach x in cycle ebike walk {
gen wkmmets_`x' = mmet_`x' * ((cyc_dist_km*cyclecommute_tripspertypicalweek)/speed_`x') // mMETS OF CYCLING PER WEEK IN HOURS AMONG NEW CYCLISTS
}
gen wkmmets_cycle_dutch=((1-percentebike_dutch)*wkmmets_cycle)+(percentebike_dutch*wkmmets_ebike)
gen wkmmets_cycle_ebike=((1-percentebike_ebike)*wkmmets_cycle)+(percentebike_ebike*wkmmets_ebike)
** MORTALITY / SICKNESS PROTECTION
foreach x in pa sick {
gen cprotection_`x'_govtarget= 1-(rr_ref_`x'^(wkmmets_cycle/mmet_ref))
gen cprotection_`x'_nocyclists=cprotection_`x'_govtarget
gen cprotection_`x'_govnearmkt=cprotection_`x'_govtarget
gen cprotection_`x'_gendereq=cprotection_`x'_govtarget
gen cprotection_`x'_dutch= 1-(rr_ref_`x'^(wkmmets_cycle_dutch/mmet_ref))
gen cprotection_`x'_ebike= 1-(rr_ref_`x'^(wkmmets_cycle_ebike/mmet_ref))
gen wprotection_`x'= 1-(rr_ref_`x'^(wkmmets_walk/mmet_ref))
}
foreach x in nocyclists govtarget govnearmkt gendereq dutch ebike {
recode cprotection_pa_`x' 0.45/max=0.45
recode cprotection_sick_`x' 0.5/max=0.5
}
recode wprotection_pa 0.3/max=0.3
recode wprotection_sick 0.5/max=0.5
** DEATHS AVOIDED + YLL
foreach x in nocyclists govtarget govnearmkt gendereq dutch ebike {
gen `x'_sic_death=`x'_sic*mortrate*cprotection_pa_`x'*-1
gen `x'_siw_death=`x'_siw*mortrate*wprotection_pa*-1
gen `x'_sideath=`x'_sic_death+`x'_siw_death // here and for CO2, not 'gen long' as individual
drop `x'_sic_death `x'_siw_death
}
gen base_sldeath = -1 * nocyclists_sideath // BASELINE LEVEL IS INVERSE OF 'NO CYCLISTS' SCENARIO INCREASE
gen base_slyll = base_sldeath * yll_per_death_discounted
gen base_slvalueyll = base_slyll * vsly * -1
foreach x in nocyclists govtarget govnearmkt gendereq dutch ebike {
gen `x'_sldeath=`x'_sideath+base_sldeath
gen `x'_slyll=`x'_sldeath * yll_per_death_discounted
gen `x'_siyll=`x'_sideath * yll_per_death_discounted
gen `x'_slvalueyll=`x'_slyll * vsly * -1
gen `x'_sivalueyll=`x'_siyll * vsly * -1
}
** SICKNESS ABSENCE COST
foreach x in nocyclists govtarget govnearmkt gendereq dutch ebike {
gen `x'_sic_sick=`x'_sic*sickness_hours_year*cprotection_sick_`x'
gen `x'_siw_sick=`x'_siw*sickness_hours_year*wprotection_sick
gen `x'_sisickdays=(`x'_sic_sick+`x'_siw_sick)/-7.5 // here and for CO2, not 'gen long' as individual
gen `x'_sivaluesick=`x'_sisickdays*salary_hourly*-7.5 // here and for CO2, not 'gen long' as individual
gen `x'_sivaluecomb=`x'_sivaluesick+`x'_sivalueyll // here and for CO2, not 'gen long' as individual
drop `x'_sic_sick `x'_siw_sick
}
gen base_slsickdays = -1 * nocyclists_sisickdays // BASELINE LEVEL IS INVERSE OF 'NO CYCLISTS' SCENARIO INCREASE
gen base_slvaluesick = -1 * nocyclists_sivaluesick
gen base_slvaluecomb = base_slvaluesick + base_slvalueyll
foreach x in govtarget govnearmkt gendereq dutch ebike {
gen `x'_slsickdays=`x'_sisickdays+base_slsickdays
gen `x'_slvaluesick=`x'_sivaluesick+base_slvaluesick
gen `x'_slvaluecomb=`x'_slvaluesick+`x'_slvalueyll
order `x'_slyll `x'_slvalueyll `x'_slsickdays `x'_slvaluesick `x'_slvaluecomb /*
*/ `x'_sideath `x'_siyll `x'_sivalueyll `x'_sisickdays `x'_sivaluesick `x'_sivaluecomb, after(`x'_sldeath)
}
order base_sldeath base_slyll base_slvalueyll base_slsickdays base_slvaluesick base_slvaluecomb, after(ebike_sicyclehours)
** CO2
foreach x in nocyclists govtarget govnearmkt gendereq dutch ebike {
gen `x'_sicartrips =`x'_sid * cyclecommute_tripsperweek * 52.2 // NO. CYCLISTS * COMMUTE PER DAY
gen `x'_sicarkm=`x'_sid * cyclecommute_tripsperweek * 52.2 * cyc_dist_km // NO. CAR TRIPS * DIST
gen `x'_sico2 = `x'_sicarkm * co2kg_km
}
gen base_slcartrips=-1*nocyclists_sicartrips // BASELINE LEVEL IS INVERSE OF 'NO CYCLISTS' SCENARIO INCREASE
gen base_slcarkm=-1*nocyclists_sicarkm
gen base_slco2=-1*nocyclists_sico2
order base_slcartrips base_slcarkm base_slco2, before(govtarget_sicartrips)
foreach x in govtarget govnearmkt gendereq dutch ebike {
gen `x'_slco2=`x'_sico2+base_slco2
order `x'_sicartrips `x'_sicarkm `x'_slco2 , before(`x'_sico2)
}
** SAVE
drop mortrate sickness_hours_year salary_hourly nocyclists* gradient-hrs_ebike wkmmets_cycle - wprotection_sick
foreach x in govtarget govnearmkt gendereq dutch ebike {
foreach y in death yll valueyll sickdays valuesick valuecomb co2{
drop `x'_sl`y'
}
}
compress
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\CommuteSP_individual.dta", replace
***********************************
** SET GEOGRAPHY [ONCE FOR EACH]
***********************************
global geography = "lsoa" // "msoa" or "lsoa"
*****************
** PART pre5: PREPARE INDIVID FOR AGGREGATION [LSOA AND MSOA]
*****************
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\CommuteSP_individual.dta", clear
* PREPARE FOR AGGREGATION
rename home_$geography geo_code_o
rename work_$geography geo_code_d
gen all=1
gen bicycle=(commute_mainmode9==1)
gen foot=(commute_mainmode9==2)
gen car_driver=(commute_mainmode9==3)
gen car_passenger=(commute_mainmode9==4)
gen motorbike=(commute_mainmode9==5)
gen train_tube=(commute_mainmode9==6)
gen bus=(commute_mainmode9==7)
gen taxi_other=(commute_mainmode9==8)
drop home*soa work*soa home_lad11cd home_laname home_gor
drop commute_mainmode9- incomedecile cyc_dist_km cyc_avslope_perc *cartrips
* MAKE BIDIRECTIONAL OD AND SAVE TEMPORARY DATASET, PRE-AGGREGATION
gen osub1=substr(geo_code_o,1,1)
gen dsub1=substr(geo_code_d,1,1)
gen osub2=substr(geo_code_o,2,.)
gen dsub2=substr(geo_code_d,2,.)
destring osub2 dsub2, replace force
gen homefirst=.
recode homefirst .=1 if (osub1=="E" & dsub1!="E")
recode homefirst .=1 if (osub1=="W" & dsub1!="E" & dsub1!="W")
recode homefirst .=0 if (osub1=="W" & dsub1=="E")
recode homefirst .=1 if (osub1=="E" & dsub1=="E" & osub2<= dsub2)
recode homefirst .=1 if (osub1=="W" & dsub1=="W" & osub2<= dsub2)
recode homefirst .=0 if (osub1=="E" & dsub1=="E" & osub2>dsub2)
recode homefirst .=0 if (osub1=="W" & dsub1=="W" & osub2>dsub2)
gen geo_code1=geo_code_o
replace geo_code1=geo_code_d if homefirst==0
gen geo_code2=geo_code_d
replace geo_code2=geo_code_o if homefirst==0
drop osub1 dsub1 osub2 dsub2 homefirst
/* MAKE DATASET OF GORDET [ONLY NEED HASHED OUT BIT ONCE NOT EACH TIME]
preserve
keep geo_code_o home_gordet
rename geo_code_o geo_code
rename home_gordet gordet
duplicates drop
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\geo_code_gordet.dta", replace
restore
*/
rename home_gordet gordet_o
rename geo_code1 geo_code
merge m:1 geo_code using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\geo_code_gordet.dta", nogen
rename geo_code geo_code1
rename gordet gordet_1
* IDENTIFY VARIABLES WHERE 'ALL' IS TOO SMALL (for lsoa combine <3 flows to 'under 3' for flowtype 1 and 2)
bysort geo_code1 geo_code2: gen f_all_temp=_N
replace geo_code_d="Under 3" if f_all_temp<3 & "$geography"=="lsoa" & flowtype<=2
replace gordet_1=gordet_o if geo_code_d=="Under 3"
replace geo_code1=geo_code_o if geo_code_d=="Under 3"
replace geo_code2=geo_code_d if geo_code_d=="Under 3"
drop f_all_temp flowtype
gen id = geo_code1+" "+geo_code2
order id geo_code1 geo_code2 geo_code_o geo_code_d gordet_o gordet_1 all- taxi_other
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\CommuteSP_preaggregate_temp.dta", replace
*****************
** PART 5A: AGGREGATE TO ZONE LEVEL [LSOA AND MSOA]
*****************
forval reg=1/11 {
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\CommuteSP_preaggregate_temp.dta", clear
* RESTRICT TO HOME REGION
keep if gordet_o==`reg'
* AGGREGATE UP AREA FIGURES
foreach var of varlist all- taxi_other govtarget_slc- ebike_sico2 {
bysort geo_code_o: egen a_`var'=sum(`var')
}
* PERCENT TRIPS AND TRIP HILLINESS
recode rf_dist_km min/9.9999=1 10/max=0, gen(rf_u10km_dist) // NB for msoa layer this is actually based on LSOA distances
recode rf_u10km_dist .=0 if geo_code_d=="Other"
* NB keep as missing if no fixed work place - implicitly assume they have same distribution as everyone else. This exclusion is comparable to what ONS do
bysort geo_code_o: egen a_perc_rf_dist_u10km=mean(rf_u10km_dist*100)
gen rf_u10km_avslope_perc=rf_avslope_perc
replace rf_u10km_avslope_perc=. if rf_u10km_dist!=1
bysort geo_code_o: egen a_avslope_perc_u10km=mean(rf_u10km_avslope_perc)
* AREA FILE KEEP/RENAME + MERGE IN NAMES/LA + ORDER
keep geo_code_o a_*
rename geo_code_o geo_code
rename a_* *
duplicates drop
merge 1:1 geo_code using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\geo_code_lookup.dta"
drop if _m==2
drop _m
order geo_code geo_name lad11cd lad_name all bicycle- taxi_other
* SAVE + APPEND
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\z_all_reg`reg'.dta", replace
}
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\z_all_reg1.dta", replace
forval reg=2/11 {
append using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\z_all_reg`reg'.dta"
}
export delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\z_all_attributes_unrounded.csv", replace
forval reg=1/11 {
erase "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\z_all_reg`reg'.dta"
}
*****************
** PART 5B: AGGREGATE TO FLOW LEVEL [LSOA AND MSOA]
*****************
forval reg=1/11 {
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\CommuteSP_preaggregate_temp.dta", clear
* RESTRICT TO STARTING ZONE REGION
keep if gordet_1==`reg'
* AGGREGATE UP FLOW FIGURES
foreach var of varlist all- taxi_other govtarget_slc- ebike_sico2 {
bysort id: egen f_`var'=sum(`var')
}
foreach var of varlist rf_dist_km rf_avslope_perc {
replace `var'=. if geo_code_d=="Under 3" // as combining multiple together
bysort id: egen f_`var'=mean(`var') // redundant for lsoa, useful for msoa
}
* FLOW FILE KEEP/RENAME + MERGE IN NAMES/LA + ORDER
keep id geo_code1 geo_code2 f_*
rename f_* *
duplicates drop
forval i=1/2 {
rename geo_code`i' geo_code
merge m:1 geo_code using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\geo_code_lookup.dta"
drop if _m==2
drop _m
foreach x in geo_code geo_name lad11cd lad_name {
rename `x' `x'`i'
}
}
foreach x in geo_name2 lad11cd2 lad_name2 {
replace `x'="No fixed place" if geo_code2=="OD0000003"
replace `x'="Other" if geo_code2=="Other"
replace `x'="Under 3" if geo_code2=="Under 3"
}
order id geo_code1 geo_code2 geo_name1 geo_name2 lad11cd1 lad11cd2 lad_name1 lad_name2 all bicycle- taxi_other govtarget_slc- ebike_sico2
* MERGE IN OTHER CS DATA (BETWEEN-LINES ONLY)
rename rf_dist_km rf_dist_km_1 // identical for LSOA, different for MSOA
rename rf_avslope_perc rf_avslope_perc_1
merge 1:1 id using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\rfrq_all_data.dta"
drop if _m==2
recode e_dist_km .=0 if geo_code1==geo_code2 // within-zone
replace rf_dist_km=rf_dist_km_1 if rf_dist_km==. // within-zone OR rf_dist just >30km so not in MSOA: use LSOA average here
replace rf_avslope_perc=rf_avslope_perc_1 if rf_avslope_perc==. // within-zone OR rf_dist just >30km so not in MSOA: use LSOA average here
count if _m!=3 & geo_code1!=geo_code2 & geo_code2!="Other" & geo_code2!="OD0000003" & geo_code2!="Under 3" // should be none for LSOA
count if _m!=3 & geo_code1!=geo_code2 & geo_code2!="Other" & geo_code2!="OD0000003" & rf_dist_km!=rf_dist_km_1 // should be none for MSOA
drop _m rf_dist_km_1 rf_avslope_perc_1
* SAVE + APPEND
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\od_all_reg`reg'.dta", replace
}
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\od_all_reg1.dta", replace
forval reg=2/11 {
append using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\od_all_reg`reg'.dta"
}
codebook id // check no duplicates
export delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\od_all_attributes_unrounded.csv", replace
forval reg=1/11 {
erase "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\\$geography\od_all_reg`reg'.dta"
}
* NB some 150 MSOA routes wirh rf_dist<20km, but missing rq_dist because the rf_dist<20km exists from LSOA level but the MSOA distance was >30km
*****************
** PART 5C: AGGREGATE TO LA & PCT REGION LEVEL [LSOA ONLY]
*****************
** LA
import delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lsoa\z_all_attributes_unrounded.csv", clear
* AGGREGATE
foreach var of varlist all- taxi_other govtarget_slc- ebike_sico2 {
bysort lad11cd: egen a_`var'=sum(`var')
}
foreach var of varlist perc_rf_dist_u10km avslope_perc_u10km {
bysort lad11cd: egen temp_`var'=sum(`var'*all)
gen a_`var'=temp_`var'/a_all
}
keep lad11cd lad_name a_*
rename a_* *
duplicates drop
* CHANGE UNITS
foreach x in base_sl govtarget_si govnearmkt_si gendereq_si dutch_si ebike_si {
replace `x'valueyll=`x'valueyll/1000000 // convert to millions of pounds
replace `x'valuesick=`x'valuesick/1000000 // convert to millions of pounds
replace `x'valuecomb=`x'valuecomb/1000000 // convert to millions of pounds
replace `x'cyclehours=`x'cyclehours/1000 // convert to thousands hours
replace `x'carkm=`x'carkm/1000 // convert to thousands km
replace `x'co2=`x'co2/1000 // convert to tonnes
}
export delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lad_all_attributes_unrounded.csv", replace
** REGION
import delimited "pct-inputs\01_raw\01_geographies\pct_regions\pct_regions_lad_lookup.csv", varnames(1) clear
save "pct-inputs\02_intermediate\x_temporary_files\scenario_building\pct_regions_lad_lookup.dta", replace
import delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lad_all_attributes_unrounded.csv", clear
merge 1:1 lad11cd using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\pct_regions_lad_lookup.dta", keepus(region_name) nogen
* AGGREGATE
foreach var of varlist all- taxi_other govtarget_slc- ebike_sico2 {
bysort region_name: egen a_`var'=sum(`var')
}
keep region_name a_*
rename a_* *
duplicates drop
* PERCENTAGES FOR INTERFACE
foreach var of varlist bicycle govtarget_slc govnearmkt_slc gendereq_slc dutch_slc ebike_slc {
gen `var'_perc=round(`var'*100/all, 1)
order `var'_perc, after(`var')
}
keep region_name *perc
list if bicycle_perc==.
drop if bicycle_perc==. // should be nothing
export delimited using "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\pct_regions_all_attributes.csv", replace
*****************
** PART 5D: FILE FOR RASTER WITH NUMBERS ROUNDED (BUT CORRECT TOTALS MAINTAINED), AND NO FILTERING OF LINES <3 [LSOA ONLY]
*****************
use "pct-inputs\02_intermediate\x_temporary_files\scenario_building\commute\lsoa\CommuteSP_preaggregate_temp.dta", clear
* SUBSET BY DISTANCE AND TO SCENARIO VARIABLES
keep if rf_dist_km<20 & geo_code1!=geo_code2
keep id bicycle *_slc
* AGGREGATE UP FLOW FIGURES
foreach var of varlist bicycle- ebike_slc {
bysort id: egen f_`var'=sum(`var')
}
keep id f_*
rename f_* *
duplicates drop
* ROUND + MOVE SOME CYCLISTS 0 TO 1, TO GET TOTAL CORRECT NUMBER
set seed 20170121
gen random=uniform()
foreach var of varlist govtarget_slc govnearmkt_slc gendereq_slc dutch_slc ebike_slc {
rename `var' `var'_orig
gen `var'=round(`var'_orig)
total `var'_orig `var'
matrix A=r(table)
di "`var': " round((100*(1-A[1,2]/A[1,1])),0.01) "%"
}
foreach var of varlist govtarget_slc govnearmkt_slc dutch_slc ebike_slc { // not needed gendereq, as similar with/without rounding
total `var'_orig `var' if `var'_orig<1.5
matrix A=r(table)
gen `var'_diff = round(A[1,1]-A[1,2]) // COUNT THE DIFFERENCE BETWEEN NO CYCLISTS ROUNDED VS NOT ROUNDED AMONG THOSE WHERE NOT ROUNDED IS <1.5
sort `var' random
gen littlen=_n
recode `var' 0=1 if littlen <=`var'_diff // ROUND SOME 0 TO 1 SO THAT TOTAL NO. <1.5 IS CORRECT
drop littlen
}
foreach var of varlist govtarget_slc govnearmkt_slc gendereq_slc dutch_slc ebike_slc {
total `var'_orig `var'