Re-run 2020 Wisconsin Congressional Districts

[christopherkenny]

Redistricting requirements

In Wisconsin, districts must:

1. have equal populations

Interpretation of requirements

We enforce a maximum population deviation of 0.5%. We add a pseudo-county constraint as described below.

Data Sources

Data for Wisconsin 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 Wisconsin across 2 independent runs of the SMC algorithm. We use a pseudo-county constraint to limit the county and municipality splits. Municipality lines are used in Milwaukee County. These are larger than the target population for a district. No special techniques were needed to produce the sample.

Validation

SMC: 5,000 sampled plans of 8 districts on 7,059 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0

Plan diversity 80% range: 0.59 to 0.85

R-hat values for summary statistics:
   pop_overlap      total_vap       plan_dev      comp_edge    comp_polsby       pop_hisp      pop_white      pop_black       pop_aian 
      1.018045       1.000334       1.017433       1.012493       1.001198       1.004935       1.003655       1.003070       1.000654 
     pop_asian       pop_nhpi      pop_other        pop_two       vap_hisp      vap_white      vap_black       vap_aian      vap_asian 
      1.016204       1.008083       1.000661       1.023666       1.005044       1.006400       1.002300       1.001295       1.015057 
      vap_nhpi      vap_other        vap_two pre_16_rep_tru pre_16_dem_cli uss_16_rep_joh uss_16_dem_fei uss_18_rep_vuk uss_18_dem_bal 
      1.003765       1.001014       1.024333       1.017555       1.008775       1.013189       1.021412       1.014944       1.016234 
gov_18_rep_wal gov_18_dem_eve atg_18_rep_sch atg_18_dem_kau sos_18_rep_sch sos_18_dem_laf pre_20_dem_bid pre_20_rep_tru         arv_16 
      1.011614       1.018596       1.014212       1.017597       1.017621       1.015334       1.008323       1.013519       1.015066 
        adv_16         arv_18         adv_18         arv_20         adv_20  county_splits    muni_splits            ndv            nrv 
      1.016569       1.014496       1.017223       1.013519       1.008323       1.002334       1.000965       1.015644       1.014567 
       ndshare          e_dvs         pr_dem          e_dem          pbias           egap 
      1.014480       1.013279       1.012151       1.012959       0.999987       1.009228 

Sampling diagnostics for SMC run 1 of 2 (2,500 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     2,456 (98.2%)     18.9%        0.27 1,591 (101%)     10 
Split 2     2,432 (97.3%)     25.6%        0.33 1,579 (100%)      6 
Split 3     2,389 (95.6%)     31.6%        0.45 1,550 ( 98%)      4 
Split 4     2,358 (94.3%)     32.8%        0.50 1,553 ( 98%)      3 
Split 5     2,317 (92.7%)     34.5%        0.51 1,521 ( 96%)      2 
Split 6     2,305 (92.2%)     29.6%        0.53 1,484 ( 94%)      2 
Split 7     2,311 (92.5%)     10.2%        0.50 1,316 ( 83%)      2 
Resample    1,721 (68.8%)       NA%        0.52 1,460 ( 92%)     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     2,456 (98.2%)     12.7%        0.27 1,553 ( 98%)     15 
Split 2     2,430 (97.2%)     19.2%        0.33 1,560 ( 99%)      9 
Split 3     2,405 (96.2%)     24.1%        0.43 1,568 ( 99%)      6 
Split 4     2,339 (93.6%)     28.0%        0.50 1,560 ( 99%)      4 
Split 5     2,281 (91.2%)     30.0%        0.53 1,548 ( 98%)      3 
Split 6     2,307 (92.3%)     11.2%        0.53 1,448 ( 92%)      8 
Split 7     2,302 (92.1%)      6.3%        0.54 1,326 ( 84%)      5 
Resample    1,748 (69.9%)       NA%        0.54 1,456 ( 92%)     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.

Checklist

• [x] I have followed the instructions
• [x] I have updated the tracker
• [x] All TODO lines from the template code have been removed
• [x] I have merged in the master branch and then recalculated summary statistics
• [x] I have run enforce_style() to format my code
• [x] The documentation copied above is up-to-date
• [x] There are no data files in this pull request
• [x] None of the file output paths (for the redist_map and redist_plans objects, and summary statistics) have been edited

@CoryMcCartan

Note: Consistent with previous discussions and other states, we remove the cores requirement, as it is not a state requirement, rather a court directive for how to approach the new lines.