Visibility graphs can make money in financial markets

Battle Count: 72

Highly relevant to quantitative trading. The paper directly proposes a novel technical indicator (VGRSI) for generating trading signals, demonstrates its application in an automated trading system with concrete profit results, and evaluates it using standard quantitative metrics (Sharpe ratio, drawdown, trading intensity). The walk-forward testing methodology and inclusion of transaction costs align with quantitative trading best practices. The geometric/network-based approach offers a genuinely novel perspective beyond classical trend-following indicators. However, the limited number of instruments tested and absence of comparison baselines slightly reduce the immediate practical applicability.

Moderate complexity. The core visibility graph computation (checking line-of-sight between points) is algorithmically straightforward but requires careful implementation for efficiency, especially with large windows (WS, WV up to 200). The multi-timeframe analysis (M1, M5, M30) and walk-forward optimization add engineering complexity. The indicator itself is computationally simple (no ML training required), but the full trading system with parameter grids, position management, SL/TP logic, and multi-asset coordination requires substantial implementation effort. The MQL5/Python implementation on MetaTrader 5 adds platform-specific complexity.

The methodology is described in mathematical detail with clear formulas for visibility relations, aggregation, and normalization. However, the implementation relies on MetaTrader 5 platform with MQL5 scripts (not publicly shared), and the specific parameter grids tested are only partially described. The walk-forward scheme and threshold ranges are specified, but exact code is not provided. The author reserves rights to the VGRSI concept. Reproduction would require access to the same historical data via MetaTrader 5 and reimplementation of the indicator logic.