d05 aggregate Ruhela: Age 50+ -> custodial deaths

Usage

morie_otis_analyze_d05_ruhela_aggregate(data, out_dir = NULL)

morie_otis_analyze_d05_mrm_aggregate(data, out_dir = NULL)

Arguments

data: d05 data.frame.
out_dir: Optional.

Value

morie_otis_analysis_result.

Examples

# \donttest{
otis_d05 <- morie_synth_otis("d05", n = 120L, seed = 1L)
morie_otis_analyze_d05_ruhela_aggregate(otis_d05)
#> OTIS d05 -- Aggregate RF: Age 50+ -> Number of custodial deaths
#> ===============================================================
#>   Source file                            d05_deaths_in_custody_age_category.csv
#>   Dataset id                             d05
#>   Treatment column                       T_50plus
#>   Outcome (count)                        Number_CustodialDeaths
#>   Year FE                                Year
#>   Covariate FEs                          none
#>   Cluster group (GEE)                    --
#>   Cells                                  120
#>   Total outcome count                    5101
#>   Overdispersion (Poisson AIC - NB AIC)  1105.4
#> 
#> Aggregate Ruhela formulation -- Poisson + NB IRR for treatment effect on count outcome:
#>   Family   Coefficient        IRR     95% CI          p-value   AIC     Notes
#>   -----------------------------------------------------------------------------------------
#>   Poisson  factor(T_50plus)1  1.1392  [1.071, 1.211]  3.08e-05  2224.4  n=120, total_y=5101
#>   NB       factor(T_50plus)1  1.1492  [0.884, 1.494]  2.99e-01  1119    n=120, total_y=5101
#> 
#> Aggregate RF on d05: age-50+ overrepresentation in custodial death counts; well-documented old-age mortality in custody.
#> 
#> IRR > 1 ==> treatment increases the count rate; IRR < 1 ==> treatment decreases it. Concordance between Poisson and NB indicates equidispersion; large gap (Poisson AIC > NB AIC) indicates overdispersion -- trust NB. 
# }