Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

시간적 계층을 이용한 교통사고 발생건수 예측Temporal hierarchical forecasting with an application to traffic accident counts

Authors
Jun, GwanyoungSeong, Byeongchan
Issue Date
Apr-2018
Publisher
KOREAN STATISTICAL SOC
Keywords
temporal hierarchies; reconciliation forecast; weighted least square estimator; ARIMA model; exponential smoothing method
Citation
KOREAN JOURNAL OF APPLIED STATISTICS, v.31, no.2, pp 229 - 239
Pages
11
Journal Title
KOREAN JOURNAL OF APPLIED STATISTICS
Volume
31
Number
2
Start Page
229
End Page
239
URI
https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/2307
DOI
10.5351/KJAS.2018.31.2.229
ISSN
1225-066X
2383-5818
Abstract
본 논문에서는 시간적 계층 개념을 활용하여 시계열 자료를 예측하는 방법을 소개한다. 횡단적 계층 자료 분석에서와 유사한 방법으로 중복되지 않는 시간적 계층을 시계열 자료에 구조화할 수 있다. 이러한 시간적 계층을 활용하여 조정된 예측은 기존의 계층별 독립적 기저 예측 및 상향식 예측보다 더 정확하고 강건한 예측값을 생성한다. 실증 분석으로서 국내 교통사고 발생건수를 시간적 계층 개념을 활용하여 예측한다. 분석 결과, 조정 예측이 기존의 다른 예측보다 예측 성능면에서 더 우수함을 확인할 수 있다.
This paper introduces how to adopt the concept of temporal hierarchies to forecast time series data. Similarly as in hierarchical cross-sectional data, temporal hierarchies can be constructed for any time series data by means of non-overlapping temporal aggregation. Reconciliation forecasts with temporal hierarchies result in more accurate and robust forecasts when compared with the independent base and bottom-up forecasts. As an empirical example, we forecast traffic accident counts with temporal hierarchies and observe that reconciliation forecasts are superior to the base and bottom-up forecasts in terms of forecast accuracy.
Files in This Item
There are no files associated with this item.
Appears in
Collections
College of Business & Economics > Department of Applied Statistics > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Seong, Byeong Chan photo

Seong, Byeong Chan
경영경제대학 (응용통계학과)
Read more

Altmetrics

Total Views & Downloads

BROWSE