PinnedInDataDrivenInvestorbyYasanji Ratnaike·Oct 23From option quotes to a calibrated volatility surface, and a P&L attributed to the greeksWhat a free option feed can and cannot support: cleaning, surface calibration, and a daily P&L split by greek.
PinnedInDataDrivenInvestorbyYasanji Ratnaike·Sep 27From Pairs Trading to a Portfolio in Python, Part 1: Building the EngineWhat decides a pairs trade is the difference between correlation and cointegration. This first entry builds the engine that tells the two…
PinnedInDataDrivenInvestorbyYasanji Ratnaike·Oct 7From Pairs Trading to a Portfolio in Python, Part 2: Testing the BookPart 1 built an engine for finding cointegrated pairs. This entry takes it forward through time, builds a risk framework around it, and…
InDataDrivenInvestorbyYasanji Ratnaike·Nov 10How Far Can a Team of AI Agents Run on Their Own?Eight AI agents run a $1M paper book. Here’s how it’s built, and what I learned.
InDataDrivenInvestorbyYasanji Ratnaike·Nov 2Don’t Let the AI Pick the Trades. Let It Run the Book.The rules decide every trade; the model only supervises, and the most it can do is halt the book.
InDataDrivenInvestorbyYasanji Ratnaike·Oct 7From Pairs Trading to a Portfolio in Python, Part 3: Adding to the BookPart 2 found a small edge in large-cap pairs, too thin and too concentrated to be considered reliable. This entry tests three ways of…