Forecasting Impact

Forecasting Impact

By Forecasting ImpactBusinessEducationHow ToNon-Profit
Download on the App Store

Forecasting Impact episodes

  • Stephan Kolassa, Bahman Rostami-Tabar, and Enno Siemsen

    In this episode, we host three scientists, Dr. Stephan Kolassa, Dr. Bahman Rostami-Tabar, and Prof. Enno Siemsen. They are the authors of "Demand Forecasting for Executives and Professionals." In this episode, we delve into discussions about their book.

    We discuss their motivations for writing this unique guide for professionals and the need for such a book. We explore the pivotal role of forecasting in business decisions and unpack key principles and methodologies. Our conversation navigates through causal models, stressing the importance of understanding the forecasting process. We highlight the profound impact of AI, machine learning, and human judgment in forecasting, considering cognitive biases and the potential of large language models.

    The interview continues with discussions about the integration of demand forecasting in organizations, noting ethical considerations, reasons behind forecasting failures, and common hurdles encountered in evaluating forecast quality. We conclude by providing resource recommendations for further exploration of the topic and advice for executives eager to enhance their demand forecasting skills.

    You can learn more about their book at https://dfep.netlify.app/ and order at https://www.routledge.com/Demand-Forecasting-for-Executives-and-Professionals/Kolassa-Rostami-Tabar-Siemsen/p/book/9781032507729.

    54 min
  • Priyanga Dilini Talagala and Thiyanga Talagala

    In this episode, we delve into the inspiring academic journey of two sisters from the heart of Sri Lanka. Dr Priyanga Dilini Talagala and Dr Thiyanga Talagala. Dr Priyanga is a Senior Lecturer at the University of Moratuwa, specialising in statistical machine learning and data mining, with a fervent commitment to open-source software for reproducible research. Dr Thiyanga is a Senior Lecturer at the University of Sri Jayewardenepura, focusing on large-scale time series forecasting, data visualization, and machine learning interpretability methods.  

    We had a delightful discussion and learned about their backgrounds, their transition to and from Australia, and their perspectives on the evolving landscape of forecasting in Sri Lanka. We also discuss the status of academic collaboration with industry, the unique facets of Sri Lanka's higher education system, and the role of cultural and societal dynamics in academic communities. They shared their mentoring work and availability for forecasting opportunities in Sri Lanka. 

    45 min
  • George Athanasopoulos - President of International Institute of Forecasters

    In this episode, we hosted Professor George Athanasopoulos, President of the International Institute of Forecasters (IIF) and Head of the Department of Econometrics and Business Statistics at Monash University. 

    George gave an overview of the IIF's current plans and new initiatives, including the Practitioners chapter, publishing papers in the International Journal of Applied Forecasting, the forecasting distinguished lecture series, and plans for the International Symposium on Forecasting, among others. 

    He also shared his career experience in forecasting and how he has grown in the field to his current position. George also mentioned to his research experience in hierarchical forecasting, his teaching success, and the recent stories on co-authoring and translating the famous book "Forecasting: Principles and Practice." into Greek.

    George recommends the following books and papers as influential in his career:

    • Forecasting: Principles and Practice, RJ Hyndman, G Athanasopoulos,  Otext.
    • Tsay, R. S. (1991) "Two canonical forms for vector ARMA processes." Statistica Sinica 1, 247–69.
    • Hyndman, R., Koehler, A. B., Ord, J. K., & Snyder, R. D. (2008). "Forecasting with exponential smoothing: the state space approach." Springer Science & Business Media.
    • Panagiotelis, A., Athanasopoulos, G., Gamakumara, P., & Hyndman, R. J. (2021). "Forecast reconciliation: A geometric view with new insights on bias correction." International Journal of Forecasting, 37(1), 343-359
    51 min
  • Forecasting Software Panel

    In this episode, we had the honour of having three guests on our panel: Prof Rob Hyndman, Professor of Statistics from Monash University, Federico Garaz, CTO and co-founder of Nixtla, and Eric Stellwagen, CEO and Co-founder of Business Forecast Systems.  

    We discussed a range of topics on the role of software in forecasting, the latest status, and future trends in forecasting software. The panel shed light on the importance of incorporating operational information in software and integrating decision information in the software.  We discussed some of the challenges in implementing forecasting software and getting them to work.

    The panel shared insights and tips for excelling in forecasting software. Eric recommended defining what you want to accomplish, and the needs that you want to fill before choosing any software. Fede recommended Nixtla as a resource on various forecasting software and Forecasting Principles and Practices by Rob Hyndman and George Athanasopoulos as a reference book. Rob recommended books by Hadley Wickham as great resources.

    51 min
  • 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. 
    40 min
  • Scott Cunningham, on Technological Forecasting and Social Change

    Scott Cunningham is Professor of Public Policy at the University of Strathclyde and is the Editor-in-Chief of the journal, Technological Forecasting & Social Change. 

    In this episode, we talked about technological forecasting and social change. Prof. Cunningham gave an overview of how technological forecasting, policy, and business are interwoven, and how a systematic view is important in predicting the long-term pattern in technology. He described the broader context of tech mining, and why it is important to have mid to long-term forecasts. 

    Recommendations for books and papers: 

    The Book of Why, by Dana Mackenzie and Judea Pearl
    (Paper) Vehicle Ownership and Income Growth, Worldwide: 1960-2030 by Joyce Dargay, Dermot Gately and Martin Sommer 

    45 min
  • Tao Hong, on Energy Forecasting

    In this episode, we talk to Prof. Tao Hong, a Distinguished Professor at the University of North Carolina at Charlotte. Tao provides his insights on the future of energy forecasting research, and why we need to focus on reproducibility. He discusses the Global Energy Forecasting competitions, and what we have learned from them. He also sheds light o n the importance of industry and academic collaboration, and a business model that he has implemented successfully.

    He recommends the following reading for interested readers who want to go deep into forecasting and, specifically, energy forecasting:

    Books

    • Forecasting Principles and Practice, by Rob Hyndman, and George Athanasopoulos
    • Matrix Analysis and Applied Linear Algebra by Carl D Meyer

    Paper

    • Probabilistic electric load forecasting: A tutorial review, T Hong, S Fan, International Journal of Forecasting, V 32 (3), 2016. 
    44 min
  • Galit Shmueli, on causal inference, behavioural modifications, and role of ethics.

    In this episode, we spoke to Prof Galit Shmueli, Tsing Hua Distinguished Professor at the Institute of Service Science, and Institute Director at the College of Technology Management, National Tsing Hua University. 

    Galit talked with us about the multi-disciplinary work she has done over the years, as well as the differences between statistical models that are purposed for predicting as opposed to explaining. We also discussed causal inference and how it can be used to estimate behaviour modification by the tech giants. We continued and talked about the ethics and the complexity of that landscape.  

    Galit's recommended books:  

    1.    The age of surveillance capitalism, Shoshana Zuboff
    2.     Books on causality:
         • The book of Why, Dana Mackenzie and Judea Pearl
         • Causal Inference in Statistics: A Primer, Judea Pearl, Madelyn Glymour, and Nicholas P. Jewell
         • Causality, Judea Pearl 
    3.     Mostly Harmless Econometrics: An Empiricist's Companion, Joshua D. Angrist, ‎Jörn-Steffen Pischke 

    1 hr 2 min
  • Ataman Ozyildirim, on the business cycle and leading economic indicators

    In this episode, we spoke with Dr. Ataman Ozyildrim from The Conference Board.  We discussed leading economic indicators and its importance in tracking the economy's business cycle. He provided his insights on the current situation of the economy. He continued by pointing to the changes in supply chain trends. We also talked about the digital economy and measuring innovation in organisations.  This is only a glimpse into the many excellent insights from Ataman and The Conference Board.  

    Recommended book: Business Cycles: Theory, History, Indicators, and Forecasting by Victor Zarnowitz 

    Recommended paper:  On the aggregation of probability assessments: Regularized mixtures of predictive densities for Eurozone inflation and real interest rates by FX Diebold, M Shin, B Zhang 

    41 min
  • Nikos Kourentzes, on business, AI, and hierarchical forecasting

    Nikolaos Kourentzes is Professor of Predictive Analytics at the University of Skövde (Sweden) in the Artificial Intelligence Lab as well as a member of the Centre for Marketing Analytics and Forecasting, at Lancaster University in the United Kingdom.

    In this episode, Nikos talks about the role of AI and judgement in forecasting, and what we as forecasters need to learn from other fields such as algorithmic learning. He continues with a discussion on temporal and hierarchical forecasting problems. On a more personal note, he shares with us his background, career and philosophy on the "why” behind the problems. And, finally, he discusses his new book, Principles of Business Forecasting, 2nd edition as well as recommendations for forecasting books and papers.

    35 min

About Forecasting Impact

From the publisher's feed

Forecasting Impact is a bimonthly podcast that aims to disseminate the science and practice of forecasting by introducing prominent academics, practitioners, and visionaries in the…