PyData London 2024

Navigating through financial data challenges by harnessing the power of synthetic data
06-16, 15:00–15:40 (Europe/London), Warwick

Addressing the complexities of financial time series analysis, we unravel prevalent challenges and introduce innovative solutions with the use of synthetic data. From data scarcity to imbalanced datasets, bank silos, privacy concerns and regulatory compliance, this talk dives into the multifaceted obstacles hindering effective financial modelling. Discover how synthetic data generation and augmentation emerge as pivotal solutions, offering privacy-preserving alternatives in the realm of data-driven finance. The talk will cover a broad array of synthetic data techniques ranging from simplistic to state-of-the-art models such as GANs.


In this talk, we explore the intricate landscape of financial time series analysis, shedding light on the challenges that impede progress in modeling financial data. From the scarcity of data, especially in niche markets, to the inherent imbalance in financial datasets, where rare events are underrepresented, we confront the hurdles faced by analysts and data scientists in the financial domain. Whether you're grappling with detecting fraudulent transactions, market anomalies, or devising trading strategies, this talk equips you with advanced techniques to augment your financial modeling endeavors.

In order to address some of these challenges, we introduce synthetic data generation as a transformative solution. Despite the popularity of such models in the domains of image and text generation, the application of synthetic data in finance is still in an early phase. Synthetic data offers a way to create diverse and representative datasets that mimic the characteristics of real data, empowering analysts to develop more robust and accurate models for financial time series analysis.

Outline:

A proposed outline for the structure is as follows:
1. Dive into the sensitive data concerns exacerbated by silos within banks, hindering collaboration and data sharing for analysis purposes.
2. Present ways to retrieve publicly available real-time and historical market data through a set of powerful APIs directly from Python
3. Apply various models for synthetic data generation and evaluate their respective performance.

Target Audience: This presentation is tailored for data scientists and all data enthusiasts, financial analysts, and researchers involved in quantitative finance, algorithmic trading, or financial risk management.


Prior Knowledge Expected

No previous knowledge expected

Stamatis is the Co-founder of Asappien, an application aimed at addressing post-pandemic social and digital isolation. Currently employed as a CRM Data Analyst at Aegean Airlines, he holds a bachelor's degree in finance with distinction. Pursuing a Master's in Data Science and AI at the American College of Greece, Stamatis has earned triple scholarships for his academic excellence. He is deeply passionate about traveling, mathematics, and artificial intelligence.

Elena is an Assistant Professor at Deree - The American College of Greece (ACG) and a Senior Artificial Intelligence (AI) Advisor with over 10 years of combined business, academic and lecturing experience in the fields of Machine Learning and Data Science.

Prior to her role with the American College of Greece, Elena was an Associate Director – Lead Data Scientist for the AI teams at HSBC Global Markets, UK, and Growth & Innovation at Alpha Bank, Greece. With a strong academic background in Computer Science and programming, Elena holds a PhD in Machine Learning & Data Mining, and is passionate about developing innovative solutions through data-driven strategies across various industries. She has also held several academic roles, conducting research for universities and institutes in Cambridge and London, and is a frequent speaker at machine learning conferences and events.

Elena’s work and research interests cover a broad range of AI applications with specialisation in Natural Language Processing (NLP), time series forecasting, recommender systems, and responsible AI (XAI, FAT-ML, AI Ethics), among others.

Alkiviadis is presently engaged as a Business Analyst at impruvo., an eBusiness consulting firm that specialises in consulting, recruitment, and technology services. He holds a bachelor’s degree in Marketing Communications, graduating with honours. Alkiviadis is currently pursuing a Master’s in Data Science and AI at the American College of Greece, where he has been awarded four scholarships for his academic excellence. He aims to integrate his master’s knowledge with the eCommerce industry. In addition, he harbours a passion for gadgets and succulent gardening.

Konstantinos holds a Bachelor of Science in Physics with a major in Astrophysics from the National and Kapodistrian university of Athens. He is currently pursuing a Master's degree in Data Science at the Deree College honing his skills in programming, machine learning and AI. Proficient in Python and mathematics, Konstantinos excels in analytical problem-solving and agile methodologies. In his free time, amongst other hobbies, he enjoys staying informed and acquiring knowledge on cutting-edge technology.

Data Science student and BI Analyst at Public Power Corporation, Greece. Enthusiastic about discovering insights that drive strategic decisions.