Comparative study of Manual time series components identification with automated Break for Time Series Components (BFTSC) and Group for Time Series Components (GFTSC) in Identification in Univariate Forecasting

Authors

  • Ajare Emmanuel Oloruntoba
  • Suzilah Ismail

Abstract

The main objective of this study is to compare the behavior of manual time series components identification with BFTSC (break for time series components) and GFTSC (group for time series components) identification of time series components. The weaknesses of manual time series components identification were addressed by an advanced automated method which i created from the extension of BFAST (Break for Additive, Seasonal and Trend). Manual time series components identification, BFTSC and GFTSC serve the same purpose of identifications of time series components.  BFTSC and GFTSC is considered better alternative  than the manual method  based on findings in this study  in term of precision, time period of data incubation and flexible methods. BFTSC is designed to give a combined image of all the four time series components captured in a single time plot. GFTSC is designed to capture all the time series components on a different individual point time plot. BFAST only identifies trend and seasonal components while considering all other left over components as random, identification of trend and seasonal components alone is not enough to have a clear pattern of all the time series components present in the data. BFTSC and GFTSC is created to include cyclical and irregular components and this was included in the methodology. This study uses two types of data. The first empirical data was the monthly rainfall data in Ibadan Oyo State Nigeria (2007 to 2018). The second empirical data was the seasonal data which is the quarterly sales of coca cola beverages in Benin City Edo State Nigeria (1995 to 2016). These findings indicate that BFTSC and GFTSC can provide a better alternative to Manual time series components identification technique, hence BFTSC and GFTSC is recommended for future uses.

Downloads

Published

2019-12-14

Issue

Section

Articles