These strategies represent the culmination of a dynamic internal competition between our Quantitative Trading and Asset Management departments.
Cumulative Return (%)
Volatility (%)
Sharpe Ratio
Sortino Ratio
Max Drawdown
Profit Factor
This Long-Only strategy uses the Hurst Exponent to classify Cryptocurrencies into two
behavioral regimes: Mean-Reverting and Trending. Based on this classification, for
Mean-Reverting assets, the RSI indicator is used to generate buy signals, while the MACD is
the foundation for the Trending assets.
Signals are generated daily using close prices, and target portfolio weights are assigned
uniformly among the assets with buy signals. Before rebalancing, the strategy computes the
Normalized Euclidean Distance between the current weights and the new target weights. If this
distance falls within a no-trade zone, defined by an optimal threshold δ, the strategy keeps
the current weights to avoid unnecessary turnover.
When rebalancing is triggered, it is performed through a weighted average between current
and target weights, ensuring partial adjustments rather than full reallocation. As a result, the
strategy never performs full portfolio turnover, which helps minimize transaction costs and
slippage over time.
Cumulative Return (%)
Volatility (%)
Sharpe Ratio
Sortino Ratio
Max Drawdown
Profit Factor