TSAI leverages quantitative and systematic approaches in fixed income investing

TSAI

A systematic approach to fixed income has real benefits both pre- and post-trade. Portfolios constructed using factor models (whether the portfolio is equity or multi-asset) can be optimized to achieve minimal ex ante risk, which in many cases is a more direct way to achieve the results typically expected from a stratified sampling approach.

The advantage of the former approach over the latter is that it generally results in lower tracking error because it accounts for correlations between multiple risk drivers. Ex post risk and performance attributions can be presented in a way that more closely aligns with the investment decisions taken, providing a better understanding of how the strategy performed.

As you correctly point out, the value is not in the quantity of data, but in its quality.

Until recently, and particularly as it relates to the fixed income space, the ability to extract high-quality information from large amounts of low-quality data has been limited by the capabilities of the technology. While advances in technologies such as artificial intelligence do allow us to push the boundaries and extract more high-quality information, as we do with credit curve construction for the Axioma credit spread risk model, progress can only be made if firms understand the trade-off between computing power and cost. Furthermore, these firms need to be willing to prioritize investments in more efficient, future-proof technology stacks.

The fundamental investing principle of knowing what you are investing in still holds true. Invest in factors, but make sure you understand how they are constructed. To do this, you need to really dig into the methodology and find out not only what the risk model can do, but more importantly, what it can’t do.

Data challenges in the fixed income space make it difficult for quants to develop systematic strategies similar to those we are used to in the equity markets.

For example, we recently analyzed three different ETF “systematic” strategies. We found that there were a few “smart beta” high-yield funds that said they systematically identified bonds of high quality or value by looking at the balance sheet ratios of various issuers. The question that quants, and buyers of these products, should ask themselves is: is this really a quantitative strategy, or is it just using fundamental data to identify securities with the potential for superior returns?