No manual pre-processing decisions were necessary.
Simulation Notes
We sample 5,000 districting plans for West Virginia across 4 independent runs of the SMC algorithm.
No special techniques were needed to produce the sample.
Validation
SMC: 5,000 sampled plans of 3 districts on 55 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0
ℹ Preparing WV shapefile
Plan diversity 80% range: 0.35 to 0.79
ℹ Preparing WV shapefile
R-hat values for summary statistics:
pop_overlap total_vap plan_dev comp_edge comp_polsby pop_white pop_black
1.004724 1.000832 1.000995 1.001041 1.000643 1.002544 1.001706
pop_hisp pop_aian pop_asian pop_nhpi pop_other pop_two vap_white
1.002082 1.002052 1.001623 1.001914 1.002869 1.002047 1.001106
vap_black vap_hisp vap_aian vap_asian vap_nhpi vap_other vap_two
1.002101 1.002102 1.002717 1.001718 1.002812 1.001806 1.002128
pre_16_rep_tru pre_16_dem_cli pre_20_rep_tru pre_20_dem_bid uss_18_rep_mor uss_18_dem_man uss_20_rep_cap
1.001844 1.001648 1.002676 1.001055 1.001482 1.002323 1.001800
uss_20_dem_swe gov_16_rep_col gov_16_dem_jus gov_20_rep_jus gov_20_dem_sal atg_16_rep_mor atg_16_dem_rey
1.001373 1.002377 1.001429 1.000963 1.001869 1.001814 1.001415
atg_20_rep_mor atg_20_dem_pet sos_16_rep_war sos_16_dem_ten sos_20_rep_war sos_20_dem_ten adv_16
1.001877 1.002013 1.002063 1.001841 1.001600 1.002364 1.001554
adv_18 adv_20 arv_16 arv_18 arv_20 ndv nrv
1.002323 1.001434 1.002321 1.001482 1.002170 1.003138 1.002296
ndshare e_dvs pr_dem e_dem pbias egap
1.002266 1.002608 1.002071 1.000032 1.003598 1.000091
Sampling diagnostics for SMC run 1 of 4 (1,250 samples)
Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k
Split 1 1,224 (97.9%) 2.4% 0.29 787 (100%) 4
Split 2 1,208 (96.7%) 1.0% 0.35 682 ( 86%) 3
Resample 1,085 (86.8%) NA% 0.36 1,068 (135%) NA
Sampling diagnostics for SMC run 2 of 4 (1,250 samples)
Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k
Split 1 1,223 (97.9%) 2.5% 0.30 777 ( 98%) 4
Split 2 1,219 (97.5%) 1.0% 0.30 694 ( 88%) 3
Resample 1,134 (90.7%) NA% 0.32 1,078 (136%) NA
Sampling diagnostics for SMC run 3 of 4 (1,250 samples)
Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k
Split 1 1,223 (97.9%) 2.4% 0.30 770 ( 97%) 4
Split 2 1,214 (97.1%) 1.0% 0.32 684 ( 87%) 3
Resample 1,116 (89.3%) NA% 0.34 1,083 (137%) NA
Sampling diagnostics for SMC run 4 of 4 (1,250 samples)
Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k
Split 1 1,224 (97.9%) 2.4% 0.29 784 ( 99%) 4
Split 2 1,217 (97.3%) 1.0% 0.30 696 ( 88%) 3
Resample 1,125 (90.0%) NA% 0.33 1,077 (136%) NA
• Watch out for low effective samples, very low acceptance rates (less than 1%), large std. devs. of the log
weights (more than 3 or so), and low numbers of unique plans. R-hat values for summary statistics should be
between 1 and 1.05.
CoryMcCartan
commented
1 year ago
Redistricting requirements
[In West Virginia, according to W.V. Const. art. I, § 4, districts must:
Interpretation of requirements
We enforce a maximum population deviation of 0.5%. We also merge VTDs into counties and run the simulation at the county level.
Data Sources
Data for West Virginia comes from the ALARM Project's Redistricting Data Files.
Pre-processing Notes
No manual pre-processing decisions were necessary.
Simulation Notes
We sample 5,000 districting plans for West Virginia across 4 independent runs of the SMC algorithm. No special techniques were needed to produce the sample.
Validation
Checklist
TODOlines from the template code have been removedenforce_style()to format my coderedist_mapandredist_plansobjects, and summary statistics) have been edited@CoryMcCartan