Automated trading systems may seem a bit mysterious to those who are not familiar with them. Who creates these systems, how do they test for quality, how actively do they monitor trading performance with regard to changing market conditions?
Fortunately, we were able to interview one very prolific system developer — Victor Garcia of YAtrading –whose products are featured among Algo IQ’s system list. Victor answers a number of questions that many automated system enthusiasts and investors might ask with regard to his products and process.
Before we get started, please note that the aim of this interview is to provide information to those who may find interest in automated trading. This interview is not a promotion of his systems, nor does it affirm that Halifax America agrees with or supports any of the claims. We do, however, believe in giving system developers a venue to provide relevant information. We hope you enjoy the interview.
Halifax America (HA): Tell us a little about your trading background and at how you decided to turn transform your system development skills into a professional enterprise through YAtrading?
Victor Garcia (VG): I started in the trading world, more than 20 years ago, like most investors: with a discretionary operation; buying and selling shares through the bank.
Thanks to technological advances, the banking sector evolved a lot, with the emergence of new brokers and very advanced platforms that allowed operating in complex products with much more adjusted commissions.
At that point I decided to begin evaluating my investment strategies, which I operated manually in detail, concluding that the only viable way was with strategy programming. This way I can see and analyze real behavioral statistics.
When I began programming strategies, I discovered the potential of being able to readjust the parameters in a dynamic and agile way, without human factor influence, respecting the strategy 100%.
As time went by, and seeing that the strategies were profitable, I decided to put them out to the market on a subscription basis. In addition, I gathered a team of engineers, computer scientists and financial market specialists to continually develop and evaluate different trading systems.
HA: Looking at the scale of your product offering, one thing that strikes us is the sheer number of systems you have developed. How diverse are the strategies upon comparing one with the others?
VG: In YAtrading, we have a very wide range of trend, anti-trend, and intra-daily systems. We try to include the greatest number of possibilities available in the markets, so that the investor can create systems portfolios winning in all types of markets, whether movements are up trending, down trending, or lateral.
Every published system on our website includes a description of the algorithm operation to guide the investor. If you want more information, our team will be pleased to assist you in whatever you need.
HA: Every developer has a unique approach to strategy development; something that might be considered a proprietary “edge.” What do you feel competitively differentiates your systems from other systems in the market?
VG: YAtrading has created its own method for systems development called APEO “Analysis, Programming, Evaluation and Optimization” to create and select strong trading systems and to discard systems with low mathematical expectations of success.
This helps identify that systems that have a high probability of reduced drawdowns, high profit factor, and favorable percentage of winning sessions.
HA: As prolific as you are in developing numerous systems, you must have quite a dynamic R&D (research and development) process. Tell us a bit about the process through which you develop new strategies—what you look for, how you test the quality of a strategy before it goes to market, and how you distinguish which systems “make the cut” vs. those that go back to the drawing board.
VG: The systems we develop are linked to a large number of processes. With the use of different platforms and specialized software, both trading and statistics and mathematics, we achieve the best results, and we can discard those algorithms that do not pass the filters before being published.
In YAtrading we emphasize on not over-optimizing the algorithms we develop. “Over-optimization” is a common cause of failure among developers as over-optimized systems are designed to respond to past results that never recur in the future. In contrast, our systems are designed to identify less profitable but much more consistent and repetitive processes.
Our optimization processes — with different time cuts, periods, and graphic compressions — are developed to reach the most optimal systems values.
Once the creation and optimization process is finished, we leave the systems under study between 3 and 6 months prior to publication. This way we can see if the behavior in the market in forward testing is the same as the one that occurred in back testing. It a system passes the filter, we proceed to its publication and commercialization.
HA: What are you working on now; for instance, are you developing any new systems for release in the near future?
VG: At YAtrading we are constantly investigating and working in the creation of new systems and improvements for the existing algorithms, to give a continuous and quality service to our clients.
Our objective is to have trading systems that operate in the most liquid markets and with differentiating strategies, so that clients can have highly diversified portfolios and strategies to achieve a steady yield curve.
HA: Thank you, Victor, for joining us. We wish you the best in your endeavors.