Repository navigation
AI options flow signals + QuantStats performance reporting — open source investment skills #517
tellmefrankie
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Hey QuantStats community,
Portfolio performance reporters — sharing an open source project that generates the signals QuantStats could then analyze.
AI Investment Skills:
QuantStats integration idea:
The scanner logs timestamped signals (P/C anomalies, price alerts). Run a strategy based on those signals, then analyze returns with QuantStats — sharpe, max drawdown, signal-to-noise attribution. Did filtered P/C signals outperform raw P/C signals? QuantStats would answer that.
Real signal:
XLI P/C hit 5.32 (normal: 0.5–1.2). After filtering 98% lottery volume: "extremely bearish." Raw P/C = "neutral." Backtesting that difference with QuantStats would be very revealing.
Open source:
→ https://github.com/tellmefrankie/ai-investment-skills
Anyone here using QuantStats to measure the quality of options flow signals?
All reactions