Closed christopherkenny closed 2 years ago
In Colorado, districts must:
We enforce a maximum population deviation of 0.5%.
Data for Colorado comes from the ALARM Project's 2020 Redistricting Data Files.
No manual pre-processing decisions were necessary.
We sample 5,000 districting plans for Colorado across 2 independent runs of the SMC algorithm. A partisan competitiveness constraint was used.
SMC: 5,000 sampled plans of 8 districts on 3,108 units `adapt_k_thresh`=0.985 • `seq_alpha`=0.5 `est_label_mult`=1 • `pop_temper`=0 Plan diversity 80% range: 0.46 to 0.76 R-hat values for summary statistics: pop_overlap total_vap plan_dev comp_edge comp_polsby pop_hisp 1.033023 1.012775 1.003907 1.006380 1.013037 1.009250 pop_white pop_black pop_aian pop_asian pop_nhpi pop_other 1.016050 1.015970 1.014766 1.013059 1.024864 1.006674 pop_two vap_hisp vap_white vap_black vap_aian vap_asian 1.007744 1.011416 1.016515 1.015863 1.016873 1.010805 vap_nhpi vap_other vap_two pre_16_dem_cli pre_16_rep_tru uss_16_dem_ben 1.027309 1.005074 1.007367 1.006796 1.013571 1.006939 uss_16_rep_gle gov_18_dem_pol gov_18_rep_sta atg_18_dem_wei atg_18_rep_bra sos_18_dem_gri 1.014184 1.007110 1.013171 1.006367 1.013952 1.006508 sos_18_rep_wil pre_20_dem_bid pre_20_rep_tru uss_20_dem_hic uss_20_rep_gar arv_16 1.014136 1.006460 1.013171 1.005776 1.013983 1.013733 adv_16 arv_18 adv_18 arv_20 adv_20 county_splits 1.007289 1.014109 1.006836 1.013468 1.005985 1.003699 muni_splits ndv nrv ndshare e_dvs pr_dem 1.000444 1.006130 1.013439 1.010806 1.010808 0.999802 e_dem pbias egap 1.005082 1.004414 1.005532 Sampling diagnostics for SMC run 1 of 2 (2,500 samples) Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k Split 1 1,790 (71.6%) 19.2% 0.51 1,582 (100%) 9 Split 2 1,790 (71.6%) 25.5% 0.63 1,330 ( 84%) 6 Split 3 1,783 (71.3%) 30.5% 0.67 1,327 ( 84%) 4 Split 4 1,821 (72.8%) 33.2% 0.68 1,288 ( 82%) 3 Split 5 1,837 (73.5%) 37.4% 0.69 1,321 ( 84%) 2 Split 6 1,891 (75.7%) 27.3% 0.68 1,279 ( 81%) 2 Split 7 1,780 (71.2%) 5.6% 0.68 1,200 ( 76%) 4 Resample 906 (36.2%) NA% 1.20 1,107 ( 70%) NA Sampling diagnostics for SMC run 2 of 2 (2,500 samples) Eff. samples (%) Acc. rate Log wgt. sd Max. unique Est. k Split 1 1,821 (72.9%) 12.4% 0.51 1,634 (103%) 14 Split 2 1,776 (71.0%) 19.6% 0.64 1,346 ( 85%) 8 Split 3 1,753 (70.1%) 25.4% 0.68 1,322 ( 84%) 5 Split 4 1,818 (72.7%) 25.7% 0.72 1,318 ( 83%) 4 Split 5 1,850 (74.0%) 22.7% 0.67 1,322 ( 84%) 4 Split 6 1,675 (67.0%) 17.7% 0.72 1,268 ( 80%) 4 Split 7 1,790 (71.6%) 5.7% 0.74 1,162 ( 74%) 4 Resample 961 (38.5%) NA% 1.21 1,149 ( 73%) 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.
TODO
enforce_style()
redist_map
redist_plans
delete this line and all the tags except the reviewers you need @CoryMcCartan
Competitiveness constraint comparison
Redistricting requirements
In Colorado, districts must:
Interpretation of requirements
We enforce a maximum population deviation of 0.5%.
Data Sources
Data for Colorado comes from the ALARM Project's 2020 Redistricting Data Files.
Pre-processing Notes
No manual pre-processing decisions were necessary.
Simulation Notes
We sample 5,000 districting plans for Colorado across 2 independent runs of the SMC algorithm. A partisan competitiveness constraint was used.
Validation
Checklist
TODO
lines from the template code have been removedenforce_style()
to format my coderedist_map
andredist_plans
objects, and summary statistics) have been editeddelete this line and all the tags except the reviewers you need @CoryMcCartan
Notes
Competitiveness constraint comparison![image](https://user-images.githubusercontent.com/28026893/174951723-f0e5b156-7b56-4ed7-a5a6-ec5578d9fc9b.png)