adlp: Accident and Development period Adjusted Linear Pools (ADLP)...

In ADLP: Accident and Development Period Adjusted Linear Pools for Actuarial Stochastic Reserving

Accident and Development period Adjusted Linear Pools (ADLP) Models

Description

Class to estimate an ADLP model fitted by Minorization-Maximisation.

Usage

## S3 method for class 'adlp'
print(x, ...)

adlp(components_lst, newdata, partition_func, param_tol = 1e-16, ...)

Arguments

x	Object of class adlp
...	Other named parameters passed onto further functions
components_lst	List of adlp_components
newdata	Validation data to fit the ADLP partitions on
partition_func	Partition function used to subset the data. ADLP weights will be generated for each partition. To specify partition preferences, set the parameter to adlp_partition_none if no partitioning is required. For partitioning the claims triangle by accident periods with predetermined weights, use adlp_partition_ap. Alternatively, users can create a custom partition function by defining the cut-off accident period for each subset manually.
param_tol	Tolerance for weights. Any value less than tolerance in magnitude is assumed zero.

Details

See adlp_component and adlp_components objects for more information on required format for inputs.

See adlp_partition for information on valid partition functions.

For an understanding of how partitions affect the performance of the ADLP ensemble, one might refer to Avanzi, Li, Wong and Xian (2022)

Value

Object of class adlp. This object has the following components:

components_lst

adlp_components; List of adlp_components, see also adlp_components

model_weights

vector; vector of model weights fitted for each component

partition_func

function; Partition function used to fit the components

optim_MM

mm_optim; Details related to the MM algorithm see also MM_optim()

newdata

data.frame; Data.frame used to fit the ADLP

References

Avanzi, B., Li, Y., Wong, B., & Xian, A. (2022). Ensemble distributional forecasting for insurance loss reserving. arXiv preprint arXiv:2206.08541.

Examples

data(test_adlp_component)
test_component1 <- test_adlp_component
test_component2 <- test_adlp_component
test_components <- adlp_components(
    component1 = test_component1,
    component2 = test_component2
)

newdata <- test_component1$model_train$data

test_adlp <- adlp(test_components, newdata = newdata, response_name = "claims",
    partition_func = adlp_partition_ap, tri.size = 40, size = 3)

ADLP documentation built on May 29, 2024, 2:12 a.m.