Skip to contents

Extract model data

Source: R/model_get_data.R
model_data.Rd

[Experimental]

model_data() retrieves underlying data from fitted models. It works for objects of class "glm", as well as objects produced by refitting procedures ("refitsmooth" or "refitrestricted").

The compatibility wrapper extract_model_data() remains available and is not deprecated.

extract_model_data() is kept as a compatibility alias for model_data(). It is not deprecated.

Usage

model_data(x)

extract_model_data(x)

Arguments

x

An object of class "glm", "refitsmooth", or "refitrestricted".

Value

A data.frame of class "model_data", containing the cleaned model data with additional attributes:

  • rf — names of risk factors in the model

  • offweights — weights or offsets if present

  • mgd_rst, mgd_smt — merged restrictions/smooths (refit objects only)

  • new_nm, old_nm — new and old column names (refit objects only)

Details

For GLM objects, the function:

  • returns the original data used in the model,

  • attaches attributes with the relevant rating factors and any weights/offsets.

For refit objects, the function:

  • strips out auxiliary columns used for smoothing/restrictions,

  • attaches attributes with information about rating factors, merged smooths, restrictions, and offsets.

Author

Martin Haringa

Examples

if (FALSE) { # \dontrun{
library(insurancerating)
library(dplyr)

# Fit GAM for claim frequency
age_policyholder_frequency <- riskfactor_gam(data = MTPL,
                                             nclaims = "nclaims",
                                             x = "age_policyholder",
                                             exposure = "exposure")

# Determine clusters
clusters_freq <- construct_tariff_classes(age_policyholder_frequency)

# Add clusters to MTPL portfolio
dat <- MTPL |>
mutate(age_policyholder_freq_cat = clusters_freq$tariff_classes) |>
mutate(across(where(is.character), as.factor)) |>
mutate(across(where(is.factor), ~biggest_reference(., exposure)))

# Fit frequency and severity model
freq <- glm(nclaims ~ bm + age_policyholder_freq_cat, offset = log(exposure),
            family = poisson(), data = dat)
sev <- glm(amount ~ bm + zip, weights = nclaims,
           family = Gamma(link = "log"), data = dat |> filter(amount > 0))

# Add predictions for freq and sev to data, and calculate premium
premium_df <- dat |>
add_prediction(freq, sev) |>
mutate(premium = pred_nclaims_freq * pred_amount_sev)

# Fit unrestricted model
burn_unrestricted <- glm(premium ~ zip + bm + age_policyholder_freq_cat,
weights = exposure, family = Gamma(link = "log"), data = premium_df)

# Impose smoothing and refit model
burn_restricted <- burn_unrestricted |>
add_smoothing(x_cut = "age_policyholder_freq_cat",
x_org = "age_policyholder",
breaks = seq(18, 95, 5)) |>
refit_glm()

# Extract model data
model_data(burn_restricted)
} # }