Poisson + Negative-Binomial IRR for volatility ~ alert complexity

Builds the (id x fiscal year) cell with outcome vm (number of distinct regions visited) and treatment T_high_ac = 1 if the person-year alert-complexity ac >= 2, then fits Poisson and (optionally) negative-binomial GLMs adjusting for Year, Gender, and Age. The NB fit uses MASS::glm.nb when available; if not, only Poisson is reported.

Usage

morie_otis_irr_glmm_vm(df)

Arguments

df

b01 data.frame.

Value

morie_otis_result.

Details

No random effect / cluster-robust SE – for paper-grade inference, use the dedicated OTIS DML pipeline.

Examples

b02 <- morie_synth_otis("b02", n = 120L, seed = 1L)
res <- morie_otis_irr_glmm_vm(b02)
res$title
#> [1] "IRR Poisson/NB GLM (vm)"