M conference series

The M Conferences follow the M Competitions. Their purpose is to disseminate their findings, discuss its conclusions, exchange information about its implications of forecasting practices and recommend ways to improve them in the future.



M6 Conference

The Makridakis Open Forecasting Center (MOFC) conducts conferences after the end of each M competition with the latest being “The Future of Forecasting and the M6 Competition, Conference”.

The M6 Forecasting Competition took place from February 2022 to February 2023, which brought together a diverse range of participants from around the globe to forecast financial (stock and ETF) prices and explore the relationship between forecast accuracy and investment returns.

Read the Press Release

The Future of Forecasting and the M6 Conference ran on the 6th and 7th of November 2023 at the Harvard Club of New York, in New York City, USA. This well attended conference covered many critical and emerging aspects of forecasting, including the role of AI and large language models, forecasting for the new era and the changing role of the forecaster. Furthermore, how to best measure the performance of forecasts, and the value of the findings from forecasting competitions were presented during the Conference.

The M6 competition’s success highlights the critical role of accurate forecasting and uncertainty estimation in our rapidly changing world, and its significance has been recognized by industry leaders who also served as sponsors and co-organizers. These included Google, Meta, J.P. Morgan, International Institute of Forecasters, Kinaxis, Intech, causaLens, SaS, ForecastPro, Erasmus School of Economics, Forecasting and Strategic Unit at the National Technical University of Athens Greece (NTUA), the Centre for Decision Making and Risk Analysis at INSEAD, MOFC and the University of Nicosia

Professor Spyros Makridakis

Professor Spyros Makridakis and the other organizers of the M6 competition elaborated on the winning methods, and their practical implications. The conference program also included talks from distinguished speakers from the major software/technology companies, as well as known academics from top-level universities. It featured presentations of the three most accurate methods of the M6 Competition by the developers themselves and also discussed how their methods could be implemented by others.

Many of the 63 M6 Competition winners attended the conference, some presented their winning methods and those present were given their prizes during the award ceremony.

During the award ceremony, Professor Makridakis was honored for his immerse contribution to the world of forecasting. Commemorative speeches were given by Antonis Polemitis, the CEO of the University of Nicosia; Anil Gaba from INSEAD; Chris Fry from Google and Michael Gilliland, Editor-in-Chief of the International Journal of Applied Forecasting.

Finally, the program included two panel discussions that focused on the organizational aspects of forecasting, the future of forecasting and the upcoming M7 Competition.

M5 Conference

The M5 Forecasting Competition ended on the 30 of June 2020, with an huge number of around 10,000 participants from more than 100 countries, making it one of the largest of its kind and the fourth most popular Kaggle competition ever. The sponsors and the winners got together at a virtual M5 Conference. Google, Walmart, Uber, International Institute of Forecasters (IIF), MOFC, INSEAD and NTUA provided financial or academic support. Along with the many innovations, there were substantial monetary prizes, totaling $100,000, with global tech giants Google, Walmart, and Uber, as well as IIF, financially backing the M5 competition.

M4 Conference

The findings of the M4 Competition were presented The M4 Conference took place in December 2018 in New York City with the title “Advances In Forecasting Machine Learning and Statistical Methods and a Novel Hybrid Approach”

M publications

The findings, conclusions and discussions of the M competitions are being published in academic and professional journals aimed at disseminating available knowledge and advancing the theory and practice of forecasting with the M competition papers having received more than 7,500 citations so far while motivating the publishing of more than 100 papers written, expanding and corroborating the M competitions. There has also been the paper “A brief history of forecasting competitions” stating: Forecasting competitions are now so widespread that it is often forgotten how controversial they were when first held, and how influential they have been over the years. “In addition, there has been special issues of the International Journal of Forecasting (IJF) exclusively devoted to the M competitions. These issues include many papers describing the winning methods as wells as discussion papers and commentaries of the various aspects of the corresponding competition.

A related publication to the M Competitions is the paper “The future of forecasting competitions: Design attributes and principles” published in Informs the Journal of Data Science (IJDS)

Another paper “The Impact of Large Language Models like ChatGPT on Forecasting” was published in the journal forecasting covering the crucial topic of AI. So is the co-edited book (M. Hamoudia, S. Makridakis and E. Spiliotis) “Forecasting with Artificial Intelligence: Theory and Applications”, exploring various theoretical and practical issues of AI.

The MOFC publications are well received in the forecasting literature gaining a large number of citations. For instance, the Google Scholar citations of Makridakis has exceeded the 34,500 ones raising his world and Cyprus ranking considerably.