Open topepo opened 8 months ago
With the new tidymodels offsets package (offsetreg), I realize that we have not documented that this can already be done:
library(tidymodels) library(offsetreg) library(poissonreg) us_deaths$offset <- log(us_deaths$population) fit <- poisson_reg() %>% fit(deaths ~ gender + age_group + year + offset(offset), data = us_deaths) tidy(fit) #> # A tibble: 9 × 5 #> term estimate std.error statistic p.value #> <chr> <dbl> <dbl> <dbl> <dbl> #> 1 (Intercept) -18.3 0.136 -135. 0 #> 2 genderMale 0.328 0.000388 845. 0 #> 3 age_group35-44 0.443 0.00177 250. 0 #> 4 age_group45-54 1.21 0.00155 780. 0 #> 5 age_group55-64 1.99 0.00146 1367. 0 #> 6 age_group65-74 2.71 0.00143 1899. 0 #> 7 age_group75-84 3.65 0.00141 2579. 0 #> 8 age_group85+ 4.77 0.00140 3402. 0 #> 9 year 0.00568 0.0000672 84.5 0``` <sup>Created on 2024-01-23 with [reprex v2.0.2](https://reprex.tidyverse.org)</sup>
With the new tidymodels offsets package (offsetreg), I realize that we have not documented that this can already be done: