View source: R/model_get_data.R
construct_model_points | R Documentation |
construct_model_points()
is used to construct model points from generalized linear models, and must
be preceded by model_data()
. construct_model_points()
can also be used
in combination with a data.frame.
construct_model_points( x, exposure = NULL, exposure_by = NULL, agg_cols = NULL, drop_na = FALSE )
x |
Object of class model_data or of class data.frame |
exposure |
column with exposure |
exposure_by |
split column exposure by (e.g. year) |
agg_cols |
list of columns to aggregate (sum) by, e.g. number of claims |
drop_na |
drop na values (default to FALSE) |
data.frame
Martin Haringa
## Not run: # With data.frame library(dplyr) mtcars %>% select(cyl, vs) %>% construct_model_points() mtcars %>% select(cyl, vs, disp) %>% construct_model_points(exposure = disp) mtcars %>% select(cyl, vs, disp, gear) %>% construct_model_points(exposure = disp, exposure_by = gear) mtcars %>% select(cyl, vs, disp, gear, mpg) %>% construct_model_points(exposure = disp, exposure_by = gear, agg_cols = list(mpg)) # With glm library(datasets) data1 <- warpbreaks %>% mutate(jaar = c(rep(2000, 10), rep(2010, 44))) %>% mutate(exposure = 1) %>% mutate(nclaims = 2) pmodel <- glm(breaks ~ wool + tension, data1, offset = log(exposure), family = poisson(link = "log")) model_data(pmodel) %>% construct_model_points() model_data(pmodel) %>% construct_model_points(agg_cols = list(nclaims)) model_data(pmodel) %>% construct_model_points(exposure = exposure, exposure_by = jaar) %>% add_prediction(., pmodel) ## End(Not run)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.