Classifies algorithmic trading ZigZag patterns using a comprehensive taxonomy of two-pivot and three-pivot swing structures.
This skill provides a rigorous framework for identifying and labeling ZigZag swing patterns in financial time-series data. It categorizes price movements into exhaustive variants—including Higher Low (HL), Equal Low (EL), and Lower Low (LL)—while accounting for volatility through epsilon tolerance bands and Freedman-Diaconis binning. Ideal for quant developers and traders, it maps structural patterns to specific market regimes like trend continuation, triangle compression, and rally failures to enhance algorithmic decision-making and pattern recognition accuracy.
주요 기능
01Freedman-Diaconis (FD) binning for granular swing sub-classification
029-variant and 27-variant exhaustive pattern classification models
03Market regime mapping for bullish, bearish, neutral, and transition states
04Taxonomy for two-pivot (UP-DOWN) and three-pivot (UP-DOWN-UP) ZigZag patterns
0540 GitHub stars
06Dynamic epsilon tolerance band calculations for volatility-adjusted price equality
사용 사례
01Implementing standardized price action logic in trading bots and financial pipelines
02Automating pattern labeling for quantitative trading datasets and backtesting
03Identifying market regime shifts based on structural swing changes