A Framework for Predictive Analysis of Stock Market Indices : A Study of the Indian Auto Sector

TL;DR

This paper analyzes the Indian Auto sector stock index from 2010 to 2015, decomposes its components, and proposes five forecasting methods, demonstrating their effectiveness in handling complex time series data.

Contribution

It introduces five novel forecasting approaches based on structural analysis of stock index components, improving prediction accuracy in volatile market conditions.

Findings

Decomposition of stock index into trend, seasonal, and random components.

Proposed forecasting methods outperform existing techniques.

Effective prediction despite high volatility and trend changes.

Abstract

Analysis and prediction of stock market time series data has attracted considerable interest from the research community over the last decade. Rapid development and evolution of sophisticated algorithms for statistical analysis of time series data, and availability of high-performance hardware has made it possible to process and analyze high volume stock market time series data effectively, in real-time. Among many other important characteristics and behavior of such data, forecasting is an area which has witnessed considerable focus. In this work, we have used time series of the index values of the Auto sector in India during January 2010 to December 2015 for a deeper understanding of the behavior of its three constituent components, e.g., the trend, the seasonal component, and the random component. Based on this structural analysis, we have also designed five approaches for forecasting and also computed their accuracy in prediction using suitably chosen training and test data sets. Extensive results are presented to demonstrate the effectiveness of our proposed decomposition approaches of time series and the efficiency of our forecasting techniques, even in presence of a random component and a sharply changing trend component in the time-series.