Open AccessModelling of fault in RPM using the GLARMA and INGARCH model
Author(s) -
Kim JiYong,
Kim HeeYoung,
Park Daihee,
Chung Yongwha
Publication year - 2018
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
eISSN - 1350-911X
pISSN - 0013-5194
DOI - 10.1049/el.2017.3398
Subject(s) - autoregressive model , heteroscedasticity , autoregressive–moving average model , series (stratigraphy) , mathematics , star model , fault (geology) , negative binomial distribution , poisson distribution , moving average , setar , generalized linear model , econometrics , statistics , time series , autoregressive integrated moving average , paleontology , seismology , biology , geology
According to the of time series of faults in railway point machines (RPMs), forecasting approach based on the generalised linear autoregressive moving average (GLARMA) models and the integer‐valued generalised autoregressive conditional heteroscedastic (INGARCH) models are presented. The conditional distribution of observed fault counts of given previous faults and weather conditions are assumed to be Poisson or negative binomial distributions. The forecasting future fault counts of RPM are obtained by one‐step‐ahead forecasts and the performance evaluation shows that the GLARMA method performs better than the traditional autoregressive moving‐average (ARMA) model and generalised linear model (GLM).