Evaluating Indicator Independence for VGT Trading with Rank and Distance Correlation
Summary
The document proposes a framework for selecting indicator pairs for trading the Vanguard Information Technology ETF (VGT). It argues that correlated indicators can create redundant confirmation, particularly in a concentrated technology ETF with frequent price gaps and changing volatility. The proposed approach uses Kendall’s Tau and distance correlation to assess how independently candidate indicators behave, then uses selected combinations to form trading signals. Candidate tools discussed include trend, momentum, volatility, and exhaustion measures.
The article situates the analysis in seasonal and market context, including possible fourth-quarter strength, year-end volatility expansion, and subsequent rotation or post-earnings pressure. It presents this context as motivation for evaluating indicator combinations, not as a guaranteed seasonal effect. The supplied excerpt describes a custom MQL5 signal class and concludes that statistical dependence measures can help reduce duplicated inputs. However, it provides limited detail here on the actual pair selection, data construction, correlation results, or strategy performance. Its seasonal observations and proposed signals therefore require independent testing, especially given ETF gaps, concentration, and changing market regimes.
Key ideas
- The article uses Kendall’s Tau and distance correlation to evaluate dependence between candidate indicators.
- Reducing redundant indicators may prevent a system from treating repeated information as independent confirmation.
- VGT’s technology concentration and price gaps are presented as relevant differences from continuously traded currency pairs.
- Seasonal and volatility observations motivate the analysis but do not establish reliable predictive effects.
- The excerpt offers limited detail on pair selection results and trading performance, so independent testing remains necessary.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.