On the 24th November, we welcomed students from various universities based within the UK to our ‘Machine Learning in Finance: Time Series forecasting in Python’ workshop, enabling students to learn from experts about applying Data Science techniques to real-world data — specifically focusing on Time Series forecasting — thus increasing the number of skills they leave university with.
Throughout the day, students had the opportunity to network with various leaders from the HSBC Global Banking and Markets and Cambridge Spark teams — enabling them to widen their professional network and learn more about starting their career in Data Science and Engineering.
Introducing students to Machine Learning and Time Series Forecasting
We started the day with an introduction to the HSBC Global Banking and Markets team, where they went further into their backgrounds, what they do in their roles, before sharing their valuable professional experiences with the students.
And we're off! We're starting the session with a presentation by the @HSBC_UK Global Banking and Markets team, talking about their team, opportunities and experiences. pic.twitter.com/w7utxmNl6V
— Cambridge Spark (@CambridgeSpark) November 24, 2018
Following the introduction, Dr. Raoul-Gabriel Urma, CEO at Cambridge Spark, kicked off the Machine Learning in Finance session, before handing off to our expert tutors, Ricardo and Alan who led the core modules; immersing students deep in the theory, before they practiced and honed-in on their newly-found skills and knowledge in a series of Jupyter notebooks.
Over the course of the day, we covered:
- Dealing with time
- Time series autoregression
- Autoregressive–moving-average (ARMA) models
- Time Series testing
- Facebook Prophet
These modules are taught in our Linear Models and Time Series Forecasting short course
The Competitive Hackaton
Upon covering the modules, Kevin, CTO at Cambridge Spark, introduced the concluding competition, whereby students used real-world smart meter data to create models which predicted the future energy consumption of consumers, before submitting their hard word to K.A.T.E.® — our proprietary AI-powered training and assessment platform which students originally used to complete an exercise on to compete for a place on the workshop.
It's competition time! Now we've wrapped up the @HSBC_UK workshop, students are working with real-world data from smart meters to create models which predict future energy consumption, before submitting them on K.A.T.E.®. pic.twitter.com/zLTIgf8KW8
— Cambridge Spark (@CambridgeSpark) November 24, 2018
And we have our winners! Meet Johannes, Sofiane, Adit, and Thanusan — students from Imperial College London and Warwick University.
The team of four managed to obtain the highest score in the competition, before presenting their work to the rest of the group; enabling them to share the winning formula which helped secure them the winning position.
Congratulations to the winning team of our Time Series Forecasting competition at the @HSBC_UK workshop! Keep a lookout for our upcoming article next week, outlining our interview with them. pic.twitter.com/tcI9G3kdYL
— Cambridge Spark (@CambridgeSpark) November 24, 2018
We spoke to the winners at the end of the competition to learn about their experience of the day and the competition. Here’s what they had to say…
Yohannes, Masters student at Imperial College London:
“For me, what was really great about the day, was seeing a variety of approaches to time series financial data — which was excellent. Particularly about the competition at the end, what was great about that was that we used a novel way to go about data analysis using Facebook Prophet.”“…This was particularly great as we got to see how some of the leading tech companies are building APIs, and other libraries, that anyone can use to get some really great insights.”
Thanusan, recent graduate from the University of Warwick:
“The competition was really great in the aspect that it really allows you to emulate what you do in real life in a sense that you have to use real techniques, such as having a training set, having to test it and learning how to cross-validate…”“…The day was really useful in terms of learning what professionals in industry do, plus it was really great having that bit of competition with other students attending too.”
Concluding the day
Following the day-long event, students were accompanied by the HSBC and Cambridge Spark teams, where students were able to learn more about what it takes to kick-start their careers and succeed within HSBC’s Data Science and Engineering departments first-hand from leaders at the global banking institution.
It was great to see so many students using cutting-edge tools and technologies to interact with financial data while exploring real-world scenarios and AI-powered platforms. Congratulations to all the winners and thanks to all of those who participated. We look forward to greater engagement in 2019."
- Matthew Sattler, Head of Business Insights & Analytics, HSBC
Interested in attracting technical talent?
Whether you're looking to hire early careers or experienced technical talent, we have a range of innovative solutions to enable you to expand your reach and attract highly-engaged high-calibre talent.
Please complete the form to the right of this text with your details, along with a brief description of your requirements. We'll then get in touch and guide you through every step of the way.
Get in touch now
Please complete all of the required fields to get in touch with us.
-1.png)
Enquire now
Fill out the following form and we’ll contact you within one business day to discuss and answer any questions you have about the programme. We look forward to speaking with you.
