Adjust the intercept to regain the balance property

This function updates the intercept term of the model fit such that the balance property is satisfied.

adjustIntercept(obj, data)

Arguments

obj

an object of type glm, cpglm or cpglmm containing the model fit.

data

a data.frame or data.table object that was used to fit the model.

Value

The object with the adjusted (fixed effects) coefficients.

References

Campo, B.D.C. and Antonio, Katrien (2023). Insurance pricing with hierarchically structured data an illustration with a workers' compensation insurance portfolio. Scandinavian Actuarial Journal, doi: 10.1080/03461238.2022.2161413

Wüthrich, M. V. (2020). Bias regularization in neural network models for general insurance pricing. European actuarial journal 10(1), 179–202.

Examples

library(statmod)
datas  = dataCar[1:1e3, ]
Fit    = glm(Y ~ area + gender, data = datas, weights = datas$w, family = tweedie(1.75, 0),
model = TRUE, control = glm.control(epsilon = 1e-4, maxit = 5e2))
w      = weights(Fit, "prior")
y      = Fit$y
sum(w * y) == sum(w * fitted(Fit))
#> [1] FALSE
adjFit = adjustIntercept(Fit, datas)
coef(adjFit)
#> (Intercept)       areaB       areaC       areaD       areaE       areaF 
#>   5.5386226  -0.1882927  -0.1542479  -2.4340072  -1.6156363   0.8378340 
#>     genderM 
#>   0.2379400 
sum(w * y) == sum(w * fitted(adjFit))
#> [1] FALSE