Forecasting Time Series

"I have seen the future, and it is just like the present, only longer."  This quote from "The Profit" (a parody of "The Prophet") actually contains subtle insight into statistical forecasting.  It's not a magic wand - it is simply a collection of tools for projecting into the future what we think we know about how the present works.  Statistical forecasting is both an art and a science - the science part is covered in Galit Shmueli's "Forecasting Time Series," offered online at statistics.com. For more details please visit at http://www.statistics.com/forecasting.

Dr. Shmueli is Professor of Data Analytics at the Indian School of Business (SRITNE) in Hyderabad, and Associate Professor of Statistics in the department of Decision, Operations & Information Technologies at the Smith School of Business, University of Maryland. Dr. Shmueli's research has been published in the statistics, information systems, and marketing literature; she is a co-author of "Data Mining for Business Intelligence," "Modeling Online Auctions," and "Statistical Methods in e-Commerce Research."

Aim of Course:

This course will teach you how to choose an appropriate time series forecasting method, fit the model, evaluate its performance, and use it for forecasting. The course will focus on the most popular business forecasting methods: Regression models, smoothing methods including Moving Average (MA) and Exponential Smoothing, and Autoregressive (AR) models. It will also discuss enhancements such as second-layer models and ensembles, and various issues encountered in practice.

Who Should Take This Course:

Business analysts, sales forecasters, economists, financial analysts, anyone who needs to produce, interpret or assess forecasts will find this course useful. Participants should be familiar with basic statistics, including linear regression.

Course Program:

Course outline: The course is structured as follows

SESSION 1: Characterizing Time Series and the Forecasting Goal; Evaluating Predictive Accuracy and Data Partitioning

• Visualizing time series
• Time series components
• Forecasting vs. explanation
• Evaluating predictive accuracy

SESSION 2: Regression-Based Models

• Capturing trends with linear regression
• Capturing seasonality with linear regression
• Measuring and interpreting autocorrelation
• Evaluating predictability and the Random Walk
• Second-layer models using Autoregressive (AR) models

SESSION 3: Smoothing-Based Methods

• Model-driven vs. data-driven methods
• Centered and training Moving Average (MA)
• Exponential Smoothing (simple, double, triple)
• Differencing
• ARMA and ARIMA models
• Estimation of models

 

SESSION 4: Forecasting in Practice

• Improving forecasts via ensembles
• Multiple seasonal patterns
• Automated forecasting (series partitioning, changing behavior, missing values)
• Handling managerial forecast corrections

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune (www.c-elt.com).

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com