DeepTriangle: A Deep Learning Approach to Loss Reserving

We propose a novel approach for loss reserving based on deep neural networks. The approach allows for joint modeling of paid losses and claims outstanding, and incorporation of heterogeneous inputs. We validate the models on loss reserving data across lines of business, and show that they improve on...

Full description

Bibliographic Details
Main Author:	Kevin Kuo
Format:	Article
Language:	English
Published:	MDPI AG 2019-09-01
Series:	Risks
Subjects:	loss reserving machine learning neural networks
Online Access:	https://www.mdpi.com/2227-9091/7/3/97

id	doaj-9bc6adab42be4c3d9a4a5183e90c670f
record_format	Article
spelling	doaj-9bc6adab42be4c3d9a4a5183e90c670f2020-11-24T22:21:01ZengMDPI AGRisks2227-90912019-09-01739710.3390/risks7030097risks7030097DeepTriangle: A Deep Learning Approach to Loss ReservingKevin Kuo0Kasa AI, 3040 78th Ave SE #1271, Mercer Island, WA 98040, USAWe propose a novel approach for loss reserving based on deep neural networks. The approach allows for joint modeling of paid losses and claims outstanding, and incorporation of heterogeneous inputs. We validate the models on loss reserving data across lines of business, and show that they improve on the predictive accuracy of existing stochastic methods. The models require minimal feature engineering and expert input, and can be automated to produce forecasts more frequently than manual workflows.https://www.mdpi.com/2227-9091/7/3/97loss reservingmachine learningneural networks
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Kevin Kuo
spellingShingle	Kevin Kuo DeepTriangle: A Deep Learning Approach to Loss Reserving Risks loss reserving machine learning neural networks
author_facet	Kevin Kuo
author_sort	Kevin Kuo
title	DeepTriangle: A Deep Learning Approach to Loss Reserving
title_short	DeepTriangle: A Deep Learning Approach to Loss Reserving
title_full	DeepTriangle: A Deep Learning Approach to Loss Reserving
title_fullStr	DeepTriangle: A Deep Learning Approach to Loss Reserving
title_full_unstemmed	DeepTriangle: A Deep Learning Approach to Loss Reserving
title_sort	deeptriangle: a deep learning approach to loss reserving
publisher	MDPI AG
series	Risks
issn	2227-9091
publishDate	2019-09-01
description	We propose a novel approach for loss reserving based on deep neural networks. The approach allows for joint modeling of paid losses and claims outstanding, and incorporation of heterogeneous inputs. We validate the models on loss reserving data across lines of business, and show that they improve on the predictive accuracy of existing stochastic methods. The models require minimal feature engineering and expert input, and can be automated to produce forecasts more frequently than manual workflows.
topic	loss reserving machine learning neural networks
url	https://www.mdpi.com/2227-9091/7/3/97
work_keys_str_mv	AT kevinkuo deeptriangleadeeplearningapproachtolossreserving
_version_	1725772731837841408

DeepTriangle: A Deep Learning Approach to Loss Reserving

Similar Items