Originally this interview with Alex Krishtop, founder of Edgesense, was taken by Paul Golden to be published as part of an article in Euromoney. Below we publish the full version of the interview.
Paul: How would you define quantamental analysis and is it a widely understood term?
Alex: It is not a term yet, at least not like we understand terms in, say, mathematics. However this is a convenient word to denote any synergy between quantitative and fundamental approaches to market analysis, investment and trading. Thus, it may mean adding formality to fundamental research, or adding some fundamental perspective to quantitative models.
The former meaning, which is the most acknowledged today, is typically widespread among investors, as they have always been the main employers of fundamental data, and their decisions have always been based mostly on qualitative market analysis. So, “quantamental” for them sounds mostly like replacing a condition “buy the most performing stocks” with “buy if the stock’s price momentum is greater than a certain value”. Adding some formality usually helps making these decisions more precise and less discretionary.
Quantitative traders typically use only time series data, often intentionally disregarding fundamental factors. But there’s a serious pitfall in this approach: for example, similarly looking chart patterns may be caused by totally different fundamental factors, and vice versa, the same factor may cause visually different patterns. Thus pure quantitative models start to underperform if they were developed only using patterns, and then the pattern falls apart because in fact it was caused by totally different market processes.
I have always developed quantitative strategies based on trading ideas derived from certain real market process as opposed to abstract mathematical concepts or machine learning. Therefore when the term “quantamental” emerged, I began to use it in this very sense, and noticed that so did many of quant traders. So, for me quantamental means finding a trading idea, suggesting a qualitative description for it and then building a quantitative model which reflects this qualitative description.
Paul: How do fundamental factors such as changes in the market structure affect the performance of buy-side FX strategies?
Alex: This kind of fundamental factors are perhaps the most influential and yet the most overlooked by many traders, especially quantitative. The reason is that changes in the market structure are relatively slow and you can’t notice them in a chart in the same way you can do it with, say, consolidated market reaction to an important news release. Besides that it’s really very difficult to formalize factors related to market structure to use as numbers, thus most of models disregard them.
However if you disregard these factors during the development stage, then as the market is undergone changes in structure and regime, the performance degrades, but without understanding the reason of this degradation the trader may have really hard time to decide when to stop – or whether to stop at all. Thus both discretionary and quant traders continue to trade old models which are no longer adequate to the changed market – with quite a predictable result.
One of the most notable examples could be the period of market restructuring caused by increased transparency requirements, after Dodd-Frank. Many quant models died because they were only fit to certain patterns which either disappeared or even started to work in the opposite way because of these transformations.
Paul: What are the benefits to combining quantitative and fundamental analyses to inform trades, for example in FX markets?
Alex: In my opinion the quantamental approach in the sense discussed above could be the best solution to the problem of underperformance during “transitional periods” from one market regime to another. The most reasonable and efficient way of improving the performance would be to moderate a quantitative model using a qualitative analysis of various fundamental factors.
Besides that using the quantamental approach you may expect more stable results with quantitative models compared to those designed using traditional technical analysis or machine learning.
For example strategies which exploit rebalancing of currency portfolios by large players or various short-term liquidity-related regularly repeating processes in the long run will inevitably outperform various “black boxes”, when the developer sometimes never knows why his strategy works, even on paper – a typical case for machine learning, for example. Besides that I believe that really the key benefit of using fundamental analysis in quant trading is the ability to avoid significant and long-term losses during key shifts in market regimes.