Re-run 2020 Mississippi Congressional Districts

[christopherkenny]

Redistricting requirements

In Mississippi, under Mississippi Code 5-3-123 and Committee agreement, districts must:

1. be contiguous
2. have equal populations
3. be geographically compact
4. preserve county and municipality boundaries as much as possible
5. comply with the Voting Rights Act of 1965

Interpretation of requirements

We enforce a maximum population deviation of 0.5%. We ensure that there is a majority minority district with at least 55% VAP.

Data Sources

Data for Mississippi 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 Mississippi, across two independent runs of the SMC algorithm. We apply a hinge Gibbs constraint of strength 25 to encourage drawing a majority black district.

Validation

MC: 5,000 sampled plans of 4 districts on 1,834 units
`adapt_k_thresh`=0.985 • `seq_alpha`=0.5
`est_label_mult`=1 • `pop_temper`=0

Plan diversity 80% range: 0.33 to 0.65

R-hat values for summary statistics:
    pop_overlap       total_vap        plan_dev       comp_edge     comp_polsby        pop_hisp 
      1.0161978       1.0135821       1.0017838       1.0059392       1.0131546       1.0231597 
      pop_white       pop_black        pop_aian       pop_asian        pop_nhpi       pop_other 
      1.0046979       1.0001948       1.0195683       1.0105299       1.0098101       1.0225349 
        pop_two        vap_hisp       vap_white       vap_black        vap_aian       vap_asian 
      1.0308993       1.0091235       1.0040708       1.0006973       1.0219826       1.0121113 
       vap_nhpi       vap_other         vap_two  pre_16_rep_tru  pre_16_dem_cli  uss_18_rep_wic 
      1.0173080       1.0161155       1.0123284       1.0010476       1.0022840       1.0029217 
 uss_18_dem_bar uss_r18_rep_hyd uss_r18_dem_esp  pre_20_rep_tru  pre_20_dem_bid  uss_20_rep_hyd 
      1.0036974       1.0024034       1.0032830       1.0021244       1.0028947       1.0014444 
 uss_20_dem_esp          arv_16          adv_16          arv_18          adv_18          arv_20 
      1.0043998       1.0010476       1.0022840       1.0029217       1.0036974       1.0016682 
         adv_20   county_splits     muni_splits             ndv             nrv         ndshare 
      1.0030305       1.0025706       1.0197658       1.0060115       1.0013998       1.0071866 
          e_dvs           e_dem           pbias            egap 
      1.0071928       0.9999397       1.0004984       1.0205619 

Sampling diagnostics for SMC run 1 of 2 (2,500 samples)
         Eff. samples (%) Acc. rate Log wgt. sd  Max. unique Est. k 
Split 1     1,893 (75.7%)     11.8%        0.33 1,560 ( 99%)      9 
Split 2     1,362 (54.5%)     17.1%        0.75 1,458 ( 92%)      5 
Split 3     1,270 (50.8%)      9.2%        1.12 1,186 ( 75%)      3 
Resample      490 (19.6%)       NA%        3.08   967 ( 61%)     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     1,884 (75.3%)      9.0%        0.32 1,587 (100%)     12 
Split 2     1,413 (56.5%)     12.8%        0.68 1,440 ( 91%)      7 
Split 3     1,105 (44.2%)      6.9%        1.02 1,246 ( 79%)      4 
Resample      300 (12.0%)       NA%        3.09   855 ( 54%)     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

[christopherkenny]

To confirm in advance, all sampled plans have 1 performant district, except 5 plans which have 2.

[christopherkenny]