Asset And Portfolio Management

Big data

Commerzbank Quantitative Asset Management (QAM) strategies are founded on ideas originating from quantitative fields often found outside the traditional financial sphere. As a result the team applies scientific methods and advanced investment technology to its investment strategies. These strategies are systematic, which means that investment decisions are driven by carefully programmed computers executing millions of operations on millions of data items.

Our approach to big data

The analysis of colossal quantities of data, or ‘big data’, is enabled by the industrial strength infrastructure of a global bank. The ability of our quantitative researchers to harness the big data approach allows the team to profit from an ‘information arbitrage’. Information arbitrage is the ability of quantitative models to create an information edge over the traditional techniques of human managers when it comes to trading, analysis and research.

Consequently, systematic trading models can discover new patterns and trends that connect price activity to a greater breadth of variables. This enables the construction of highly intelligent portfolios, as a result of improved diversification possibilities and the discovery of new investment hypotheses. This approach to big data means that QAM can apply its investment philosophy to different geographies, markets and/or sectors, enabling a flexible approach whereby investors can have a strategy tailored according to their desired parameters.

Equity Market Neutral

One of QAM’s flagship strategies, the Equity Market Neutral (EMN) strategy, is a data-driven, systematic strategy which monitors a diverse universe comprising over half a million equity listings.

In their search for alpha, EMN strategies seek to take advantage of pricing inefficiencies in equity securities. Big data techniques not only allow our quantitative researchers to broaden the scope of variables that they study in order to identify a greater wealth of inefficiencies, but they also greatly improve the speed at which these immense data sets are processed.

Previously, QAM applied its EMN strategy to the European market; however, given the flexibility of quantitative approaches, this strategy can transcend geographic barriers. Here it will be shown that an effective EMN strategy can be quickly and successfully applied to the US market using quantitative techniques.

Portfolio construction

In creating its EMN US strategy, QAM adopted a careful screening process to reduce its initial universe of over half a million equities down to a selective universe containing only those US stocks that satisfy our minimum standards for quality and liquidity. From this universe of approximately 1,000 stocks, the team then screens thousands of different data items to identify clusters of stocks with similar characteristics called factors. The factors in the EMN strategy are carry, liquidity, momentum, quality and value. The team aims to keep a diversified portfolio of stocks, yet it also has clear limits to restrict the number of stocks in the portfolio to ensure that exposure to specific factors is not diluted.

At a portfolio level, the five factor strategies are combined using the concept of risk budgeting. The risk budgeting approach reduces exposure to the more volatile factor strategies, while increasing exposure to the more stable factor strategies. The factor portfolio forms the long component of the strategy. Market exposure is then neutralised by taking short positions in index futures, based on the market beta of the portfolio.

Delivery of the strategy

The Equity Market Neutral strategy can be tailored to an investor’s requirements. In particular, the strategy can be applied successfully to different international markets, as is shown here by the EMN strategy's application to the US market (Table 1, Chart 1).

The strategy can be delivered in a variety of wrappers, such as managed accounts for small portfolios. Investors have full transparency and a clear view of the strategy holdings over time.

Table 1: Simulated past performance of the Equity Market Neutral US strategy

Performance since inception (2003)

Annualised return


Annualised volatility


Max. drawdown


Sharpe ratio


Chart 1: Simulated past performance of the Equity Market Neutral US strategy compared to 1-3 m T-Bills
Chart 1: Simulated past performance of the Equity Market Neutral US strategy compared to 1-3 m T-Bills
Source: Commerzbank, Bloomberg