광역권 도시철도 수요의 시계열 특성 분석

이광섭, 엄진기, 문대섭 et al. · 한국철도학회 학술발표대회논문집 · 2014

Monthly transit ridership usually shows a time-series form with a seasonal pattern, a cyclical component and/or a trend. The objective of this study is to compare and analyze monthly subway ridership in different metropolitan regions, including Seoul, Busan, Daejeon, Daegu and Gwangju. There are several methods of seasonal time-series analysis such as an ARIMA (Autoregression Moving Average) model, a Winter’s method for exponential smoothing, a decomposition method. However, a decomposition method is often not straightforward especially because the method assumes that factors (e.g., trends, seasonality, cycle, irregularity) are independent each other. Because number of transit boarding may be probabilistic, this study utilizes the ARIMA model based on Box-Jenkins method and analyzes the characteristic of time-series of subway ridership in different metropolitan regions.

Cited by 0 works.

Authors