Lesson 3 of 12 · All lessons in this module
  1. Fintech and the four Vs of Big Data
  2. Sources of Big Data and its challenges
  3. Financial data science and what makes financial data hard
  4. AI and how machine learning learns
  5. Splitting data for machine learning: size, leakage and time order
  6. Overfitting, underfitting and model fit
  7. Supervised, unsupervised and deep learning
  8. Reinforcement learning and the four learning approaches compared
  9. Inside a neural network: layers, backpropagation, GANs and VAEs
  10. Data science: processing and visualisation
  11. Text analytics, NLP and data science tools
  12. Large language models and generative AI in finance

This lesson is part of the 2027 curriculum. You are following the 2026 curriculum, where it is not taught in this form. Switch if you are sitting the exam under the 2027 curriculum.

Lesson 3 of 12 · 13 min

Financial data science and what makes financial data hard

Financial data science applies data science to prices, returns, accounting and macro data, and it needs domain knowledge because financial data are fast, time-ordered, noisy, non-linear, fat-tailed and sometimes sparse, and patterns mined from them are easy to over-trust.

In short

  • Quantitative data are numbers (prices, returns, rates, accounting figures); qualitative data are non-numeric (sentiment, management quality, geopolitics). NLP and sentiment analysis turn the second into numbers.
  • Data science is the interdisciplinary field that handles and interprets big data end to end; financial data science applies it to financial data and needs domain knowledge of markets, rules and instruments.
  • High-frequency data arrive every millisecond or microsecond; high-frequency trading lives on them and needs low-latency systems.
  • Data mining looks for patterns algorithmically; without real-world context it finds noise (overfitting: training error far below test error).
  • AI and ML regulation is evolving (privacy, transparency, systemic risk). Regulated firms face strict model-validation rules; all firms face model risk.
  • Financial data: high volume and speed, time-dependent, noisy and non-stationary, non-linear and interdependent, seasonal, fat-tailed, and sometimes sparse. Data lakes and distributed computing handle the scale.

Unlock this lesson free for 7 days

Create a free account to get 7 days free: every lesson, video and flashcard, and the question bank. No card needed.

Start your 7-day free trialLog in