A Predictive Analysis of the Indian FMCG Sector using Time Series Decomposition - Based Approach

Abstract

Abstract. Stock price movements being random in its nature, prediction of stock prices using time series analysis presents a very difficult and challenging problem to the research community. However, over the last decade, due to rapid development and evolution of sophisticated algorithms for complex statistical analysis of large volume of time series data, and availability of high-performance hardware and parallel computing architecture, it has become possible to efficiently process and effectively analyze voluminous and highly diverse stock market time series data effectively, in real-time. Robust predictive models are being built for accurate forecasting of values of highly random variables such as stock price movements. This paper has presented a highly reliable and accurate forecasting framework for predicting the time series index values of the fast moving consumer goods (FMCG) sector in India. A time series decomposition approach is followed to understand the behavior of the FMCG sector time series for the period January 2010 till December 2016. Based on the structural analysis of the time series, six methods of forecast are designed. These methods are applied to predict the time series index values for the months of 2016. Extensive results are presented to demonstrate the effectiveness ofthe proposed decomposition approaches of time series and the efficiency of the six forecasting methods. Keywords. Time series decomposition, Trend, Seasonal, Random, Holt Winters Forecasting model, Auto Regression (AR), Moving Average (MA), Auto Regressive Integrated Moving Average (ARIMA), Partial Auto Correlation Function (PACF), Auto Correlation Function (ACF). JEL. G11, G14, G17, C63.