Today’s markets are not the same as 30 years ago. The impact of algorithmic trading on the markets is substantial and changes, as well as fundamental movements within the markets, are much faster than ever before. Today it is more important than ever in history to know how to adapt. And this should also be one of the features of your current ATS.
There are many ways to approach a problem with an adaptive ATS; from simple methods to very sophisticated methods (which are technically far behind the skills of a common user). However, a common user does not despair, as there are still some solid approaches to ensure further adaptation of our ATS systems within the ever-changing markets.
Today I will describe three of these methods and my experience with them. I would like to comment that these are basic approaches, accessible to any common user.
1. Adaptive indicators
A wide range of self-adaptive indicators have been around for many years. Its principle is quite simple: these indicators mostly contain one of the means of measuring the volatility or trend of the market.
A simple example of such indicators can be KAMA (Kaufman Adaptive Moving Average). There is nothing complicated about it. You just have to add one more component to a regular moving average: a component that will “calculate” where the markets are right now; whether they are currently in a trending or non-trending phase. For example, Perry Kaufman used another of his own indicators called the Efficiency Ratio (ER) for KAMA. This indicator simply fluctuates in the range 0 – 1; a market closer to number 1 is trending and one closer to 0 would be less trending. Then it is only necessary to choose a period range, for example 2 to 50. An interconnection with an ER indicator will result in a self-adaptive version of a moving average using a number higher than the set range if the ER is getting close. to 0 (if ER is at 0, EMA will use period 50). The reason is that there is too much “noise” in the market, and therefore the lower periods are very inappropriate. Or vice versa: if the market is trending, lower EMA figures will be used automatically and ER will approach figure 1 (if ER is at 1, period 2 will be used automatically). The EMA indicator periods are not set here. They are dynamically changing in the 2 – 50 range (or any other chosen range) depending on how the market moves.
In practice, such self-adaptive indicator setups look quite simple. For example, AMA has three parameters to configure.
The first parameter sets the period calculated by the ER indicator, the second and third parameters define the range of the EMA period, which will automatically adapt to the current market situation (based on the ER indicator).
In practice, everything works very well and reliably, and the indicator is really self-adaptive – it adapts the EMA figures to the current market situation without any problems.
My experience with adaptive indicators varies. Most of them are quite interesting (for example, KAMA), while some, according to my experiments, are the same as any other ordinary indicator.
This adaptive category is not bad, but it is not as functional as the second category that I use constantly.
2. Periodic reoptimization of systems
Based on my experience and with the benefit of hindsight, I find it impossible to have a “universal” combination of parameters in our system. The markets are moving and changing too fast. With the help of a good quality process, it is possible to find a truly robust combination of parameters for our system (ie, setting indicator periods, etc.), but nothing can compare to regular high-quality re-optimization.
Regular reoptimization is not complicated. Basically, after a certain pre-programmed time, it carries out a new optimization of your system to obtain new parameters that comply with the latest market developments. That is, those that are more adapted to the current environment. The regular re-optimization process can also be simulated – it’s a pretty basic thing called Walk Forward Analysis (WFA) which is possible to simulate in many programs these days.
What is WFA? It’s nothing magical or complicated. We simply take data on which we are about to perform regular re-optimization of our system. We split such data into 10 identically large segments (we will try to simulate regular optimization 10 times) and then split each segment into two parts: one smaller and one larger. Most of it, typically 70-80% of the data, will be used for optimization called In-Sample (IS). Here, we carry out a basic backtest and look for (optimize) the parameters that make our system more interesting, not only from the point of view of profitability, but also from the point of view of the stability of the capital curve. . We then take the selected parameters and test the rest of the data: 20-30% that we have not used for the primary backtest and thus for any parameter optimization. This remaining data is called Out-Of-Sample (OOS) and shows us how the system is able to constantly adapt. If the system has that capability, we also carry out regular re-optimization in live trading.
Today, personally, I re-optimize each of my systems on a regular basis, that is, each system I change with I consider self-adaptive. The process of re-optimizing and selecting an ideal period, and particularly when to do it, is very important.
3. Have a plan for when to completely shut down the system and when to start using it again
This last point may not seem relevant to the issue of adaptation, but from my experience it is. From my point of view, knowing when to turn the system off when conditions are not acceptable and when to turn it back on when we exit our drawdown is one of the highest levels of adaptability.
This task is exceptionally difficult and can be approached in many ways. From fairly complex algorithms that can tell when the system is currently not suitable for a given market and will automatically turn it off for a certain period of time, to simple rules that result from our possibilities and our common sense.
The basis for such an approach must always be a reduction. The historical drawdown is an important indicator (even if it is “just” the backtest). Its outperforming in live trading definitely indicates something important, so for example the rule to turn the system off when it exceeds the historical DD by 1.5x and to turn it back on when it reaches at least 50% of its recent drawdown again, may be critical. . way to use and test.
Regarding this, I have to mention another experience that I have: what I have not found useful at all, and what I consider to be one of the worst approaches, is to filter the equity with the help of the moving average. It means, for example, turning off the system when your equity falls below your moving average. This method is very treacherous, has many pitfalls, and simply does not work.
Surprisingly, a better use can be found in drawdown-based rules. A conservative and much better approach is the use of MC and OOS intervals.
In this article I have only “touched” on an adaptive problem from the easiest point of view, which is accessible to a common user while using approaches that I fully support, eg WFA. In my experience, it is not possible to create a good quality ATS without using some adaptive elements in our workflow. On the other hand, in regular or swing intraday ATS there is no need to use an extreme approach and re-optimize the strategy almost every day or every minute. An interval of a few months is more than enough. In any case, it is useful to constantly think about how to be as prepared as possible for the constantly changing market environment and to have tools at hand that help us adapt better and faster.