Forecasting Foreign Tourist Using Intervention Analysis On Count Time Series

9th Annual Basic Science International Conference 2019

自然科学（总论）

Atmanegara, Eviyana^1 ; Suhartono^1 ; Atok, R.M.^1

Department of Statistics, Faculty of Mathematics, Computation, and Data Science, Institut Teknologi Sepuluh Nopember, Surabaya

60111, Indonesia^1

关键词: Auto-regressive;    Discrete values;    Intervention analysis;    Negative binomial;    Time series modeling;    Time series models;    Time-series data;    Tourism promotion;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/546/5/052014/pdf
DOI  :  10.1088/1757-899X/546/5/052014

学科分类：自然科学（综合）

来源: IOP

【 摘 要 】

The foreign tourist's data are count time series data that contains the discrete value. Poisson autoregressive (AR) and Negative Binomial AR are time series models used for forecasting count data. The number of foreign tourist arrivals is influenced by the series of inputs called interventions, such as the existence of bomb terror and tourism promotion. This research aims to forecast the number of foreign tourists visiting Indonesia by nationality, 2019. The number of tourist arrivals from Bahrain and Singapore represents low count data and high count data, respectively. This work employs intervention analysis on count time series model and intervention analysis on ARIMA. Intervention on Poisson AR is the best model for forecasting the number of tourist arrivals from Bahrain and Singapore to Indonesia, 2019.