Open AccessA modified ARIMA model for forecasting chemical sales in the USA
Author(s) -
Othman Mahdi Salah,
Ghadeer Jasim Mohammed Mahdi,
Iman Ahmed Abud Al-Latif
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1879/3/032008
Subject(s) - autoregressive integrated moving average , akaike information criterion , census , sales forecasting , bayesian probability , econometrics , computer science , bayesian information criterion , time series , operations research , statistics , mathematics , machine learning , artificial intelligence , population , demography , sociology
model is derived, and the methodology is given in detail. The model is constructed depending on some measurement criteria, Akaike and Bayesian information criterion. For the new time series model, a new algorithm has been generated. The forecasting process, one and two steps ahead, is discussed in detail. Some exploratory data analysis is given in the beginning. The best model is selected based on some criteria; it is compared with some naïve models. The modified model is applied to a monthly chemical sales dataset (January 1992 to Dec 2019), where the dataset in this work has been downloaded from the United States of America census (www.census.gov). Ultimately, the forecasted sales for the next three years for chemical sales in the USA is provided.