Strategies · Library · Stocks
§3.20Hedges and structures
Alpha combos
Blend a very large number of weak trading signals into one portfolio, weighting each by its own volatility and removing overlap between them with a regression. Each is too faint to trade alone after costs; combined they can be worthwhile.
- Asset class
- Stocks
- Held for
- Days
- Data
- Needs data ARCANE lacks
- Symphony
- Cannot be built as a symphony
The rule
- Signal
- Each alpha's recent returns, their volatilities and a decorrelating regression.
- Entry and exit
- Recompute weights regularly; trade the combined holdings.
- Sizing
- Weights from regression residuals scaled by each alpha's volatility, summing to full exposure.
How it fails
- Needs large data and compute
- Overfitting among many signals
- Costs erode faint edges
Data it needs
- historical returns of many alphas
- stock-level data across thousands of names
Needs thousands of signals and stocks; ARCANE has none of this.
How ARCANE uses it
Not usable directly; Alpha could echo the idea of blending several weak views with caution.
Needs a large universe and a signal factory beyond the builder.
Source: Kakushadze and Serur (2018), 151 Trading Strategies, §3.20, p. 60 of the SSRN version. Described in our words. Hedges and structures is ARCANE’s filing, not the book’s.