c10 aggregate Ruhela: Female -> median RC days (institution GEE)

Usage

morie_otis_analyze_c10_ruhela_aggregate(data, out_dir = NULL)

morie_otis_analyze_c10_mrm_aggregate(data, out_dir = NULL)

Arguments

data: c10 data.frame.
out_dir: Optional.

Value

morie_otis_analysis_result.

Examples

# \donttest{
otis_c10 <- morie_synth_otis("c10", n = 120L, seed = 1L)
morie_otis_analyze_c10_ruhela_aggregate(otis_c10)
#> OTIS c10 -- Aggregate RF: Female -> median RC days, institution-clustered
#> =========================================================================
#>   Source file                            c10_segregation_and_restrictive_confinement_aggregate_durations_by_institution.csv
#>   Dataset id                             c10
#>   Treatment column                       T_female
#>   Outcome (count)                        TotalAggregatedDays_RestrictiveConfinement
#>   Year FE                                EndFiscalYear
#>   Covariate FEs                          Region_MostRecentPlacement
#>   Cluster group (GEE)                    Institution_MostRecentPlacement
#>   Cells                                  42
#>   Total outcome count                    1435
#>   Overdispersion (Poisson AIC - NB AIC)  415.7
#> 
#> Aggregate Ruhela formulation -- Poisson + NB IRR + GEE clustered for treatment effect on count outcome:
#>   Family                                                       Coefficient        IRR     95% CI          p-value   AIC    Notes
#>   --------------------------------------------------------------------------------------------------------------------------------------------
#>   Poisson                                                      factor(T_female)1  0.885   [0.789, 0.992]  3.66e-02  812.1  n=42, total_y=1435
#>   NB                                                           factor(T_female)1  0.8558  [0.521, 1.404]  5.38e-01  396.3  n=42, total_y=1435
#>   GEE-Poisson (cluster:Institution_MostRecentPlacement, Exch)  factor(T_female)1  0.9016  [0.587, 1.384]  6.36e-01  --     groups=22, n_obs=42
#> 
#> Aggregate RF on c10: gender disparity in median RC days.
#> 
#> 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. 
# }