Why Your Trading Data Is Wrong — Financial Data Structures Explained

Every trading chart you have ever looked at is built on time bars — and time bars are one of the worst ways to structure financial data. This episode breaks down Chapter 2 of Advances in Financial Machine Learning by Marcos López de Prado.

→ Why time bars oversample quiet periods and undersample noise
→ Tick bars: sampling by trade count, not the clock
→ Volume bars and dollar bars: measuring what actually matters
→ Tick imbalance bars: detecting informed traders in order flow
→ The CUSUM filter: sampling only sustained moves, not noise
→ The full pipeline: raw ticks → bars → CUSUM → ML-ready features

"The data structure matters more than the model." — López de Prado

#quantfinance #machinelearning #algotrading #financialdata #optionstrading #datascience #timebars #volumebars

Видео Why Your Trading Data Is Wrong — Financial Data Structures Explained канала Jaidyn