Overview

We produce stock ranking and portfolio management processes for the fund management industry.   Our processes have a quantitative bias that amalgamates and analyses a large number of financial indicators and variables using the latest computational optimisation routines.  Members can view the performance of historical and currently active trading strategies on the data page.

Major benefits

• Our heuristic processes can be applied in different financial markets
• It can be utilised to develop various types of style-based portfolios, to absolute return and market neutral long/short hedge funds.
• It can be adapted to create specialised funds such as emerging markets, sustainable investment, and socially responsible funds
•  It can be licensed in multiple financial markets internationally

Our Process

Categorization of Risk and the Alpha PLUS strategy

Our philosophy is that with sufficient information, all primary risk factors that contribute to stock returns can be identified.  It is therefore possible to commoditize risk into various categories.  Rather than try and speculate purely on unknown risk-factors that are usually attributed to an alpha of a stock once known risk is accounted for, we focus on picking which risk factors to weight in a portfolio at different times.  Sometimes, for example, stocks with greater sensitivity to price / earnings, having small market capitalisation and less sensitivity to interest rate movements may be good buy choices.  However, at other times the opposite may be true.  Essentially, the choice of which risk exposure to take requires good market timing and it is this that generates the Alpha PLUS strategy.  Our technology allows us to account for a potentially large number of risk factors and then determine which combination of factors would lead to a positive outcome for stock selection at various points in time.

Evolutionary Algorithms and Fuzzy Logic Rule Bases

Traditional quantitative analysis still relies heavily on developing an economic model that will value stocks.  Limitations include the static nature of the model, inability to account for a large number of accounting, economic and financial parameters, and the time spent analysing each stock.  Our methods are flexible, holistic and fast:

Step 1- Data Collection

Information is collected on a particular market.  This could be, for example, stocks listed as part of the Nikkei 225, or S&P 500. 

Step 2 – Identification of Risk Factors

Automated processes analyse each stock and determine how sensitive it is to a host of risk factors, varying from exposure to specific commodity prices, fixed income markets, macroeconomic conditions, international financial market behaviour and detailed accounting statistics.

Step 3 – Development of Evolving Trading Rules and Stock Ranking

Trading rules are developed to determine a ranking of stocks that should be bought and sold.  As the financial environment changes, so do the trading rules.  Even a single days trading may lead to the rules needing re-adjustment.  By using specialised evolutionary algorithms at the heart of our optimisation routines, these rules evolve over time in alignment with the financial markets.  Our technology is also based on fuzzy logic.  This means that we don’t apply ‘hard and fast’ rules to select stocks, but rather consider all stocks and how close they match our evolving trading rules. 

Step 4 – Portfolio Optimisation

The ranking is geared to pick stocks based on their suitability for inclusion in a working portfolio.  Using state-of-the-art innovations, we optimise stock weightings using higher moment analysis.  Rather than just focus on risk and returns, we examine the whole distribution of stock returns and forward-looking values to weight stock holdings.  Numerous constraints can also be placed on the portfolio construction, including the maintenance of balanced sector weights, risk-adjustments and stock holding limitations.

Step 5 - Completion

We inform our clients the positions (long or short) and weightings to hold in their portfolio.