Sanctum US

Sanctum Stock-Bond US Adaptive

The algorithmic engine applies computational algorithms to a range of fundamental, price and other data to select undervalued stocks across a risk-adjusted portfolio. 

The Adaptive Value selection criteria allocates capital in accordance with the investment policy: 

• Approximately 80 long equity positions 
• High-quality US quoted businesses with sustainable profit and cashflow; 
• Companies at attractive valuations; 
• Companies which are more able to withstand a significant market downturn. 

The investment process is sequenced according to Pre-Trade Analysis, Signal Generation, and Allocation.  

Signal Generation  

Each stock is algorithmically assessed (e.g. across Value, Growth, Volatility…) with analysis over financial and other data (including e.g. share price history, P/E ratio, Return on Capital…). For example, each stock is reviewed on: 

• Value: whether the share price appears to be lower than its true worth 
• Growth: whether its price has the potential to grow in value faster than the market average.  
• Volatility: stability based on its historical volatility.   

Stock Convictions 

Combined signal generation assessments give a final conviction. The contribution of each assessment to this is optimised on a per-sector basis using machine learning techniques, and so can vary between different stocks. This optimisation leads to a static weight representing the confidence of how each algorithm performs in a particular sector and examines the full history of data covering different market regimes.    

Bond Allocations

Our algorithm is designed to seek upside return whist limiting drawdowns. Our adaptive data-driven algorithm allocates up to 50% of the available assets under management to treasury bonds in response to market conditions. This proportion is re-generated on a monthly basis at the same time as the Financial Science US portfolio is allocated.