The Fifth Elephant 2013

An Event on Big Data and Cloud Computing

Up next

Uncovering patterns and forecasting with time series data


Pranav Modi


  1. Understand time series analysis and its applications in industry and science. Uncover patterns in data - trends, seasonality, cyclical behavior.

  2. Learn intuitive visualization techniques. Methods for noise reduction, clustering of time series using shape analysis.

  3. Catch the R ‘forecast’ package in action.


Description : A time series is a sequence of observations which are ordered in time (or space). Examples of time series data include -

Business data - demand data, sales, inventory management.

Neuroscience data - EEG, EKG

Financial data - stock prices, currencies, derivatives

Climate data - tide levels, sunspots.

You will learn how to approach time series analysis, extract patterns and make predictions that have a huge impact.


Not a prerequisite, but exposure to R will help.

Speaker bio

I work as a data scientist at a large consulting firm where we are frequently consulted on time series forecasting problems. This talk is distilled out of my experiments with time series analysis and learnings so far.

I am a functional programming enthusiast who has ventured into machine learning and data analysis. At my previous company Runa I worked on machine learning while hacking lisp! I’d be happy to share my experiences with Clojure and self-learning adventures in data analysis as well.