Detecting Change in a Volatile Curve United State Stock Market (US SM) with the Use of Automated Decomposition for Time Series Components

Abstract

The main reason for this investigation is to use manual process of identification of time series components with two types of automated decomposition for time series known as automated BFTSC (break for time series components) and, automated GFTSC (Group for time series components) in detecting change in a volatile curve united states stock market (US SM). In identification of components of time series present in the seasonal data of US stock market. The data US Stock Market was a monthly data from January 2001 until December 2018 and a total of 18 years. The stock market of US is also available as a secondary data at the DataBank of University Utara Malaysia Library. The weaknesses of BFAST were corrected by the extension of BFAST to BFTSC and GFTSC. Both were created to capture the cyclical and irregular components that were not captured by BFAST technique and it was included in the methodology of this study. BFTSC and GFTSC were considered to provide a combined image of all the four components of time series while GFTSC had additional advantage of providing equations to the components automated. Evaluation using simulation data and empirical data vindicated the accuracy of BFTSC and GFTSC based on linear trend less volatile data. They are effective and better than BFAST because it was able to identify 100% of the data with the basic four time series components monthly. Both techniques detects 99 % of the entire components in the time series data in a linear trend data.