Naive (max-simultaneous-flags + binary vm) alert -> volatility builder.

Robustness alternative to morie_otis_make_pair_alert_to_volatility_ruhela(): treatment = "max simultaneous flags across the year's rows >= 2"; outcome = "any placement row with regA != regB" (binary). Produces a different treatment marginal (~24\ encoding) and a binary rather than count outcome. Used side-by-side as the Naive arm of an RDF (Ruhela Dual Formulation).

Usage

morie_otis_make_pair_alert_to_volatility_naive(df)

Arguments

df

OTIS placement-level data.frame.

Value

Named list with data, T = "T_high_ac", Y = "Y_vm_any", covariates = c("Gender", "Age_Category", "EndFiscalYear").

Examples

# \donttest{
  df <- morie_otis_load()
  morie_otis_make_pair_alert_to_volatility_naive(df)
#> $data
#>   UniqueIndividual_ID EndFiscalYear Gender Age_Category ac_naive vm_any
#> 1   2022-SYNTH-001-RC          2022   Male     25 to 49        1      0
#> 2   2022-SYNTH-002-RC          2022 Female     18 to 24        0      0
#> 3   2023-SYNTH-003-RC          2023   Male     50 to 64        2      0
#> 4   2023-SYNTH-004-RC          2023   Male     25 to 49        0      0
#> 5   2024-SYNTH-005-RC          2024 Female     50 to 64        2      0
#>   T_high_ac Y_vm_any
#> 1         0        0
#> 2         0        0
#> 3         1        0
#> 4         0        0
#> 5         1        0
#> 
#> $T
#> [1] "T_high_ac"
#> 
#> $Y
#> [1] "Y_vm_any"
#> 
#> $covariates
#> [1] "Gender"        "Age_Category"  "EndFiscalYear"
#> 
# }