About Me
Hi, I am Marco from Italy.
I am a machine learning engineer and data scientist with broad knowledge in time
series analysis, forecasting, anomaly detection, and econometrics.
I specialized in developing and enhancing business forecasting processes and practices.
Over the years I developed and contributed to core forecasting systems in sectors
like e-commerce fashion, airline, tourism and travel. I also guided and mentored
junior colleagues in developing predictive models that enable accurate forecasting.
I am passionate about advancing the field of forecasting and fostering collaboration
between academia and industry. I do academic work in time series forecasting, and
I am a Member of the International Institute of Forecasters.
I also hold multiple certifications related to the analysis and forecasting of
time series data. Recently I have enjoyed teaching professionals and university
students about time series analysis, machine learning, programming, and statistics.
I speak Italian, English, and a little bit of French.
Have a look at my Curriculum Vitae!
Professional Experience
I am a Data Scientist with 10 years of working experience.
I started to work as Junior Data Scientist in 2017, in an Italian multinational tourism company, Uvet AMEX GBT, and its airline company Blue Panorama. There, I learnt a ton of things related to programming in production and I had the opportunity to work with a lot of different data related to business travel, pharmaceutical events, online bookings, hotels' reservations, and proprietary airline data.
In 2020, I started to work as a Data Scientist for T-Voice, a start-up company of the Triboo Group, where I worked as a Senior Consultant to develop client specific data analysis solutions, and I was responsible for the coordination of the R&D Team.
In 2021, following an M&A operation, I became a Data Scientist in Blogmeter, a company of the Shibumi Group. There I helped to integrate AI solutions into the main applications of the company and I was responsible for the coordination of the Product Team.
In 2024 I moved to Wanan Luxury as a Lead Data Scientist - Forecaster Engineer. There I led the forecasting team, developing demand forecasting solutions for thousands of e-commerce fashion products (SKUs), and training junior collegues.
Throughout all my working experience I was responsible for the
development of time series forecasting models, both for internal and
clients needs.
I developed and maintained diverse statistical and machine learning models to
predict:
- product demand at different aggregation levels to support both operational and strategic decisions,
- daily inbound calls to optimize scheduling hours,
- air passenger demand over several destinations to support aircraft planning,
- business car rental demand to support fleets management, etc.
I coordinated and contributed to the development of forecasting tools to predict
future demand for tens of thousands item-level (SKUs) time series across several
markets worldwide, and with different product hierarchies.
I also helped to integrate probabilistic methods to support more informed decisions.
Moreover, I successfully improved the approach to forecasting, resulting in a
significant reduction of errors and facilitating the faster adoption of machine
learning solutions by the business.
Finally, I provided mentorship to junior colleagues, supporting their professional
growth and development in the field of forecasting.
I am also a Ph.D. Researcher in Statistics at the University of Milano Bicocca, where I am working on academic research projects related to time series forecasting.
Teaching Experience
Since 2019 I am Adjunct Professor at the University of Milan.
I teach Statistics and Mathematical Methods in few Bachelor’s degree
courses, and Time Series Forecasting, Statistical Learning and Coding
at the Master’s degree in Data Science.
Have a look at my courses.
I also offer consultancy and training about forecasting, machine learning, and the development of business specific statistical models. Over the years, I helped several companies, like Luxottica, Mediaworld, and Blue Panorama Airlines, to improve their forecasting processes and tools.
I like teaching, indeed, since 2018 I have helped more than 300
students with their exams, thesis and projects, through private
lectures and support. I mainly teach econometrics, time series,
forecasting, programming, statistics and data science related topics.
I am a top teacher on Superprof Italia
for
Econometrics & Forecasting, and
Statistics, and
Artificial Intelligence.
Publications
I work on academic research projects related to time series forecasting. In particular, I published few papers that focus on sustainable forecasting practices and the development of high-performance forecasting systems.
- Zanotti, M. (2025). On the retraining frequency of global models in retail demand forecasting. Machine Learning with Applications, 22, 100769.
- Zanotti, M. (2025). Analyzing the retraining frequency of global forecasting models: towards more stable forecasting systems. arXiv
- Zanotti, M. (2025). The cost of ensembling: is it always worth combining? arXiv
In 2021 I also published an R package related to the topic of behavioral finance. It is the first R package allowing R users to perform several types of behavioral analysis.
The software has been presented at the useR!2021 Conference and it has been listed on the Top 40 New CRAN Packaes on August 2021.
Awards
I obtained prestigious recognitions for my activity in the field of forecasting.
-
The IIF-SAS 2025 Forecasting Research Grant for the research project “Analyzing the Retraining Frequency of Forecasting Models: Towards Accurate, Stable, and Sustainable Forecasting Systems”
-
The International Symposium on Forecasting Award for the work titled “Do Global Forecasting Models Require Frequent Retraining?" presented at the ISF 2025 Conference in Beijing, China.
Education
I did almost all my studies in Italy, in Parma and Milan. I also studied in Finland, where I lived for six months and I had amazing experiences.
I graduated cum laude in the Master’s degree in Quantitative Finance at the University of Milan, with a thesis related to time series forecasting in the airline industry: “Seasonal Analysis and Forecasting Evaluation of Seasonal Models Applied to Air Traffic Demand”.
In 2018, while working, I got a second level Master in Data Science.
In 2022, I won a fully-funded position as a Ph.D. Student of Statistics at the University of Milano Bicocca, where I obtained the title of Ph.D. in Statistics in 2026.
My research topics are time series forecasting, cross-learning, forecasting stability, anomaly detection, time series foundation models (LLMs), and sustainable AI.
I always try to keep my knowledge improving, indeed, over the years I got many certifications related to data science and forecasting.
- Certified Forecaster and Demand Planner, ISCEA
- High-Performance Time Series Forecasting, BS University
- Time Series Analysis, CAC IIT
- Deep Learning for Sequence Modelling and Times Series, Deep Learning Italia
- Python for Machine Learning and APIs, BS University
- Python for Data Science Automation, BS University
- Shiny Web Applications, BS University
- Macroeconometric Foreasting, IMF
Sports & Hobbies
I was a footballer. I played in Italian minor championships until 22. I also like reading manga, playing board games and video games, table tennis, curling and footgolf.
Since 2020, I practice kitesurf, an amazing water-board sport, and whenever I can I go kiting all around the world.
I also have a little friend named Milo, who is my companion of life.