Open AccessAutomatic Modeling Methods for Univariate Series
Author(s) -
Vı́ctor Gómez,
Agustı́n Maravall
Publication year - 2000
Publication title -
wiley series in probability and statistics
Language(s) - English
Resource type - Book series
eISSN - 1940-6347
pISSN - 1940-6517
DOI - 10.1002/9781118032978.ch7
Subject(s) - autoregressive integrated moving average , outlier , univariate , identification (biology) , computer science , box–jenkins , series (stratigraphy) , data mining , artificial intelligence , missing data , time series , machine learning , multivariate statistics , paleontology , botany , biology
In this article, a unified approach to automatic modeling for univariate series is presented. First, ARIMA models and the classical methods for fitting these models to a given time series are reviewed. Second, some objective methods for model identification are considered and some algorithmical procedures for automatic model identification are described. Third, outliers are incorporated into the model and an algorithm, for automatic model identification in the presence of outliers is proposed.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom