Chariton Christou of award-winning financial services provider Tickmill talks to Elizabeth Kesterlian
We would like to let our readers know a little bit about your data science team. So, what exactly does the Tickmill data science team do?
The data science team at Tickmill is responsible for managing the trading risk of the company at any given point in time. Every day we analyse an enormous amount of trades and data in order to extract meaningful conclusions about our risk. We’re a team of mathematicians, physicists, computer scientists and engineers and each one of us contributes a different perspective about the financial markets.
What exactly is a data scientist?
Usually, a data scientist is someone who analyses and interprets large amounts of data (few Petabytes) from different resources. They use algorithms, data mining and machine learning to come up with useful predictions.
Can you tell us a little bit about your background and how you got into data science?
I got into data science during my post-Doc. I was working for the Rosetta mission of the European Space Agency (ESA) and we were running thousands of simulations to understand the data that Rosetta was providing us with through its sensors.
What are your aims for the data science team at Tickmill?
Our aim is to create one of the most sophisticated data science teams around today. We’re building a team of experts across various backgrounds that will be able to analyse and interpret financial data, manage trading risk in a systematic manner, and even predict the markets. Our ultimate target is to be able to make accurate predictions at various frequencies using modern machine learning models.
What is the ‘Tickmill’s Forecasting Challenge’ and how did the competition come about?
Tickmill’s Forecasting Challenge is one of the few competitions in Cyprus that deals with real world data in the financial markets. We’ve invited university students to forecast (predict) a financial asset by providing them with some other inputs. Inputs are described as the pricing of other financial instruments that are available to be traded and they are correlated with the target value. We believe that inputs will help students predict the target variable.
The competition itself is the result of our work over the last few months. We’ve made painstaking efforts to ensure that our firm is always ahead of the competition with regards to data science. So, with that in mind, we wanted to assist local university students to access some incredible data while enhancing their learning experience.
The competition enables students to hone their skills in machine learning.
How would you define machine learning in a few words?
Machine learning is the ability of a machine to learn by observing historical data and extracting patterns from that data.
To be able to solve the problem, what skills/knowledge are required?
To solve the problem, one would need basic programming skills in languages such as Python, R or Julia. Beyond that, it’s creativity that one needs to solve such a complex problem.
Are participants required to have any prior knowledge of the financial markets?
We don’t expect a participant to have prior knowledge of the financial markets, even though some experience in time series forecasting could help them.
Besides the financial reward, what other benefits will the winners receive?
Taking part in our competition provides a great opportunity for winners to test and hone the skills they’ve acquired during their studies. At the same time, they’re given the chance to mingle with our Data Science team members where they can ask questions and get to know individuals working in their field of study!
What are the rules of the Tickmill’s Forecasting Challenge’ and how can you enter the competition?
The basic requirement is that participants must be enrolled in a university in Cyprus. Apart from that, another main rule is that they are not allowed to use external data. One can enter the competition by following the link for registration. Upon registering on our website, we will send the link to the competition on the Kaggle website.