Generate calendar-based trading signals from month and weekday effects.
npx skills add https://github.com/wudye/traderAssistHK --skill seasonal-wudye
Please help me install this Agent Skill. Skill: seasonal Source: https://github.com/wudye/traderAssistHK/tree/main/backend/src/skills/seasonal Command: npx skills add https://github.com/wudye/traderAssistHK --skill seasonal-wudye
It helps you generate systematic trading signals from recurring calendar effects like month-of-year and day-of-week patterns, avoiding purely discretionary timing.
Configure bullish_months and/or bearish_months, optionally enable use_weekday with bullish_weekdays and bearish_weekdays, then run the signal engine to produce a per-bar signal series for each symbol.
High-intent search queries and answers about installing and using this skill.
You can generate calendar-based trading signals by mapping bullish months to long (+1) and bearish months to short (-1) positions. This approach processes OHLCV datasets indexed by a pandas DatetimeIndex to produce integer signal series for backtesting.
Yes, you can enable a weekday overlay alongside month effects. The combined confirmation mode requires alignment between the month and weekday directional signals before opening a position, which helps reduce mismatched seasonality noise.
You test the "sell in May" anomaly by configuring bearish months in the signal engine and applying it to your OHLCV time series. The engine computes month and weekday values from the DatetimeIndex to output per-bar signals for research.
It requires an OHLCV dataset indexed by a pandas DatetimeIndex. You also need pandas and numpy installed for data handling. The engine extracts month (1-12) and weekday (0=Monday) values directly from the index to compute positions.
Standalone month signals mark bullish or bearish months directly, while combined mode adds a day-of-week overlay. Combined mode only opens long or short positions when both month and weekday signals agree, filtering out conflicting seasonal patterns.
Yes, the signal engine supports cross-asset seasonal studies. By processing any OHLCV dataset with a DatetimeIndex, it generates per-bar integer signals for each symbol, enabling broad time-pattern feature engineering and backtesting.
ebrahim-saniGenerate long, short, or neutral signals from OHLCV data using month and weekday calendar effects.
kmshihab7878Train LSTM, XGBoost, and transformer models for financial price prediction and backtesting.
agiprolabsCalculate over 130 technical analysis indicators on pandas DataFrames for cryptocurrency market data.
20YN04Generate predictive trading signals from OHLCV data using machine learning models.
