Forecasting Impact
Forecasting Impact is a monthly podcast that aims to disseminate the science and practice of forecasting alongside prominent academics and practitioners in the field. Our vision is to grow the forecasting community, foster collaboration between academia and industry, and promote scientific forecasting and good practice. We’ll discuss a variety of topics in economics, supply chain, energy, AI, data analytics, healthcare, and more.
Podcast Team: Dr. Mahdi Abolghasemi, Dr. Sevvandi Kandanaarachchi, Michał Chojnowski, Dr Laila Akhlaghi, George Boretos, Mariana Menchero, Dr. Faranak Golestaneh, Arian Sultan Khan.
Future guests: if you have something interesting on forecasting to share with our audiences, please send an email to forecastingimpact@gmail.com
Forecasting Impact
Eric Siegel on Predictive Analytics Role
Eric Siegel is a leading consultant and former Columbia University professor. He is the founder of the popular Predictive Analytics World and Deep Learning World conference series.
In this episode, Eric shares his decades of experience in predictive analytics. He discusses why ML is useful, and how predictive analytics have been used in business. Eric shares his view on prescriptive analytics, AI, and also explains uplift-modelling concepts, and why it is hard and so powerful.
Eric's Recommendations
Books:
- Competing on Analytics: Updated with a New Introduction, The New Science of Winning by Thomas H. Davenport, Jeanne G. Harris, 2017
- Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst, by Dean Abbot
- Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die by Eric Siegel
Papers:
- Sculley, David, Gary Holt, Daniel Golovin, Eugene Davydov, Todd Phillips, Dietmar Ebner, Vinay Chaudhary, Michael Young, Jean-Francois Crespo, and Dan Dennison. "Hidden technical debt in machine learning systems." Advances in neural information processing systems 28 (2015).
- Elder IV, John F. "The generalization paradox of ensembles." Journal of Computational and Graphical Statistics 12, no. 4 (2003): 853-864.