TECHNIQUES FOR ESTIMATING COMPOUND PROBABILITY DISTRIBUTION BY SIMULATING LARGE EMPIRICAL SAMPLES WITH SCALABLE PARALLEL AND DISTRIBUTED PROCESSING

摘要

Techniques for estimated compound probability distribution are described. An apparatus comprising a configuration component, perturbation component, sample generation controller, an aggregation component, a distribution fitting component, and statistics generation component. The configuration component operative to receive a compound model specification and candidate distribution definition. The perturbation component operative to generate a plurality of models from the compound model specification. The sample generation controller operative to initiate the generation of a plurality of compound model samples from each of the plurality of models. The distribution fitting component to generate parameter values for the candidate distribution definition based on the compound model samples. The statistics generation component to generate approximated aggregate statistics.

主权项

1. At least one computer-readable storage medium comprising instructions that, when executed, cause a system to:
receive, at a master node of a distributed system, a compound model specification comprising frequency models, severity models, and one or more adjustment functions; distribute the compound model specification to worker nodes of the distributed system, each of the worker nodes to at least generate a portion of samples for use in predicting compound distribution model estimates; generate one or more computer-simulated events for a plurality of units using a frequency model of the frequency models; generate a ground-up severity value for an event for a unit of the plurality of units using a severity model of the severity models, the unit selected at random from the plurality of units; determine an adjusted severity value based on applying at least one of the one or more adjustment functions to the ground-up severity value generated from the severity model of the severity models; generate an aggregate value based on the ground-up severity value and an adjusted aggregate value based on the adjusted severity value for the one or more computer-simulated events; and predict the compound distribution model estimates based on samples determined from the aggregate value and the adjusted aggregate value.