Comparative Study of Statistical Predictive Analytic Techniques
Aditya Vakaskar, Uday Joshi · IOSR Journal of Computer Engineering · 2014
Prediction is very difficult, especially if it's about the future."-NielsBohr Since the early days of mankind, man has always been fascinated by the idea of knowing future.Data is being captured at a rate never before seen in history.The retailer's goal is to translate that data into bottom line profits & Predictive analytics makes that possible.The data one captures about customers, or even consumers who interact with retail operation and don't make a purchase, is more revealing than one can think of.Customer data can provide insights on everything from large and systemic patterns of global markets, workflows, national infrastructures, and natural systems to the location, temperature, security, and condition of every item in supply chain.Predictive analytics offers access to reliable, timely information; understand customers, spot trends that drive better decisions to stay ahead in a competitive marketplace.Managers have many different decisions to make monthly, weekly, daily, sometimes even hourly.In 2012, worldwide Business Analytics software market grew 8.7% year over year with revenues reaching $34.9 billion [1] [2] and is also expecting accelerated growth which will be fueled by the quest to harness the power of big data.This paper gives Comparative Study of some of the Time Series Analytic Techniques which are the foundation blocks of Predictive Analytics.