Event series prediction via non-homogeneous Poisson process modelling

Data streams whose events occur at random arrival times rather than at the regular, tick-tock intervals of traditional time series are increasingly prevalent. Event series are continuous, irregular and often highly sparse, differing greatly in nature to the regularly sampled time series traditionall...

Full description

Bibliographic Details
Main Authors: Goulding, James, Preston, Simon P., Smith, Gavin
Format: Conference or Workshop Item
Published: 2016
Online Access:https://eprints.nottingham.ac.uk/52161/
_version_ 1848798661822119936
author Goulding, James
Preston, Simon P.
Smith, Gavin
author_facet Goulding, James
Preston, Simon P.
Smith, Gavin
author_sort Goulding, James
building Nottingham Research Data Repository
collection Online Access
description Data streams whose events occur at random arrival times rather than at the regular, tick-tock intervals of traditional time series are increasingly prevalent. Event series are continuous, irregular and often highly sparse, differing greatly in nature to the regularly sampled time series traditionally the concern of hard sciences. As mass sets of such data have become more common, so interest in predicting future events in them has grown. Yet repurposing of traditional forecasting approaches has proven ineffective, in part due to issues such as sparsity, but often due to inapplicable underpinning assumptions such as stationarity and ergodicity. In this paper we derive a principled new approach to forecasting event series that avoids such assumptions, based upon: 1. the processing of event series datasets in order to produce a parameterized mixture model of non-homogeneous Poisson processes; and 2. application of a technique called parallel forecasting that uses these processes’ rate functions to directly generate accurate temporal predictions for new query realizations. This approach uses forerunners of a stochastic process to shed light on the distribution of future events, not for themselves, but for realizations that subsequently follow in their footsteps.
first_indexed 2025-11-14T20:23:19Z
format Conference or Workshop Item
id nottingham-52161
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:23:19Z
publishDate 2016
recordtype eprints
repository_type Digital Repository
spelling nottingham-521612020-05-04T18:26:06Z https://eprints.nottingham.ac.uk/52161/ Event series prediction via non-homogeneous Poisson process modelling Goulding, James Preston, Simon P. Smith, Gavin Data streams whose events occur at random arrival times rather than at the regular, tick-tock intervals of traditional time series are increasingly prevalent. Event series are continuous, irregular and often highly sparse, differing greatly in nature to the regularly sampled time series traditionally the concern of hard sciences. As mass sets of such data have become more common, so interest in predicting future events in them has grown. Yet repurposing of traditional forecasting approaches has proven ineffective, in part due to issues such as sparsity, but often due to inapplicable underpinning assumptions such as stationarity and ergodicity. In this paper we derive a principled new approach to forecasting event series that avoids such assumptions, based upon: 1. the processing of event series datasets in order to produce a parameterized mixture model of non-homogeneous Poisson processes; and 2. application of a technique called parallel forecasting that uses these processes’ rate functions to directly generate accurate temporal predictions for new query realizations. This approach uses forerunners of a stochastic process to shed light on the distribution of future events, not for themselves, but for realizations that subsequently follow in their footsteps. 2016-12-13 Conference or Workshop Item PeerReviewed Goulding, James, Preston, Simon P. and Smith, Gavin (2016) Event series prediction via non-homogeneous Poisson process modelling. In: 2016 IEEE International Conference on Data Mining (ICDM), 12-25 Dec 2016, Barcelona., Spain. https://ieeexplore.ieee.org/document/7837840/ 10.1109/ICDM.2016.0027 10.1109/ICDM.2016.0027 10.1109/ICDM.2016.0027
spellingShingle Goulding, James
Preston, Simon P.
Smith, Gavin
Event series prediction via non-homogeneous Poisson process modelling
title Event series prediction via non-homogeneous Poisson process modelling
title_full Event series prediction via non-homogeneous Poisson process modelling
title_fullStr Event series prediction via non-homogeneous Poisson process modelling
title_full_unstemmed Event series prediction via non-homogeneous Poisson process modelling
title_short Event series prediction via non-homogeneous Poisson process modelling
title_sort event series prediction via non-homogeneous poisson process modelling
url https://eprints.nottingham.ac.uk/52161/
https://eprints.nottingham.ac.uk/52161/
https://eprints.nottingham.ac.uk/52161/