ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH

ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH

In property insurance, it is interesting to know the frequency of claims at a location observed over time. Claim is a rare event so there will be many zeros in the data. Modelling claim occurrence is important to predict the probability and frequency of the future claims. As for modelling count d...

Full description

Saved in:
Bibliographic Details
Main Author: Maudy Gabrielle, Meischke
Format: Final Project
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/39065
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
id id-itb.:39065
spelling id-itb.:390652019-06-21T14:00:09ZANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH Maudy Gabrielle, Meischke, Indonesia Final Project Count Data, Time Series, Overdispersed, ZIP, Neighborhood INSTITUT TEKNOLOGI BANDUNG https://digilib.itb.ac.id/gdl/view/39065 In property insurance, it is interesting to know the frequency of claims at a location observed over time. Claim is a rare event so there will be many zeros in the data. Modelling claim occurrence is important to predict the probability and frequency of the future claims. As for modelling count data commonly used Poisson model. However the variance of the data will be greater than the mean, and Poisson model is no longer appropriate. One alternative that can be used is Zero Inflated Poisson (ZIP) distribution. Generalized Linier Model (GLM) used to build ZIP Autoregression model that depends on frequency of previous times. Neighborhood effect is added as regressor to know the effect of occurrences frequencies in surrounding neighbors to model location. Applications of this model are used in property insurance claim frequency data of DKI Jakarta. Effect of neighborhood is added to regressor with uniform and squared inverse distance weighted. text
institution Institut Teknologi Bandung
building Institut Teknologi Bandung Library
continent Asia
country Indonesia
Indonesia
content_provider Institut Teknologi Bandung
collection Digital ITB
language Indonesia
description In property insurance, it is interesting to know the frequency of claims at a location observed over time. Claim is a rare event so there will be many zeros in the data. Modelling claim occurrence is important to predict the probability and frequency of the future claims. As for modelling count data commonly used Poisson model. However the variance of the data will be greater than the mean, and Poisson model is no longer appropriate. One alternative that can be used is Zero Inflated Poisson (ZIP) distribution. Generalized Linier Model (GLM) used to build ZIP Autoregression model that depends on frequency of previous times. Neighborhood effect is added as regressor to know the effect of occurrences frequencies in surrounding neighbors to model location. Applications of this model are used in property insurance claim frequency data of DKI Jakarta. Effect of neighborhood is added to regressor with uniform and squared inverse distance weighted.
format Final Project
author Maudy Gabrielle, Meischke,
spellingShingle Maudy Gabrielle, Meischke,
ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
author_facet Maudy Gabrielle, Meischke,
author_sort Maudy Gabrielle, Meischke,
title ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
title_short ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
title_full ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
title_fullStr ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
title_full_unstemmed ANALYSIS OF PROBABILITY FOR CLAIM EVENTS IN PROPERTY INSURANCE WITH AND WITHOUT LOCATION EFFECT THROUGH ZERO INFLATED POISSON AUTOREGRESSION APPROACH
title_sort analysis of probability for claim events in property insurance with and without location effect through zero inflated poisson autoregression approach
url https://digilib.itb.ac.id/gdl/view/39065
_version_ 1821997670246383616

Similar Items