Products & Services
We are experts in algo-development & order-execution through your broker's API
01
Algorithm Coding
The process of turning an idea, indicator, or discretionary trading system into a robust systematic trading strategy is not as easy as it may first appear. There is no question (in our minds) that it is as much art as it is science. Avoiding the common pitfalls of “curve fitting”, using bad/biased data, and unrealistic assumptions are just the beginning. Gathering meaningful lessons from historic data and generating actionable and profitable methodologies for an unknown future require a specific mindset, as well as proficiency in market infrastructure and the requisite software. We love chatting about the markets, so you’re always welcome to reach out and see if we might be a good fit.
02
Strategy Definition & Refinement
A quantified trading system takes the emotion out of trading and allows you to focus your time on research. If you have an idea that you’d like to pursue, we can help you build definitive rules that allow the strategy to be tested to see if it meets your criteria for use in live trading. If you already have a set of rules to start from, we can discuss potential ways to improve the system’s performance under different market conditions.
Even if you are not seeking to create a fully automated trading system, good research is invaluable. Having statistically valid answers to questions like “how often do these events take place?” and “what were the biggest historical risks?” can dramatically transform trader profitability (and psychology).
03
Back-testing & Optimisation
The best way to evaluate a quantified trading strategy is to subject it to rigorous back testing. Although you will often hear that “past performance is not a guarantee of future returns”, there is undeniable benefit in knowing how your system would have behaved in the past if you’d been executing it faithfully. Nevertheless, this is the tip of the iceberg of creating a robust trading methodology. Equally important, having a thorough understanding of your system’s range of “normal” behaviors will allow you to determine whether the current rough patch is par for the course or substantial enough to step away and reconsider the strategy. Check out Matt’s discussion on the basics of back-testing in a presentation that he gave for the Market Technicians Association.
04
Market Signal Scanning
A very common approach to systematic trading, short of full automation, is to create market scans and alerts for your preferred setups. Ideally these will apply to both entries and exits. Perhaps you want to generate an output file that you can submit directly to your broker for execution. It’s the age of automation – so if you are not harnessing the tools available to watch more markets, assess more opportunities and act more quickly, then you’re probably not operating efficiently enough to truly compete. Even value investors focused on fundamentals require better research tools and greater nimbleness if the goal is to always be improving.
05
Amibroker Training
If you’re the kind of person that likes to implement your own solutions, we can teach you how to use AmiBroker’s scripting language (AFL) to create indicators, explorations, back tests, and optimizations. This can be structured as a formal class for a group of people, or we can take a more ad-hoc approach for individuals who want to pick and choose their topics. We can also mentor you if you’ve already jumped into AmiBroker and just need help figuring some things out or troubleshooting a problem.
If you prefer the convenience and lower cost of recorded training sessions, you can find some of my courses at Alvarez Quant Trading and Marwood Research. Any other site selling my courses has likely stolen them, so please don’t patronize these illegal operations.
06
Integrations & Other Platforms
If you want to customise indicators or scans within TDAmeritrade’s thinkorswim platform, we can definitely help. If you’re looking at creating an Order Management System that manages all your trades through the Interactive Brokers API, we have ample experience with the technology. Need a bit of help with Excel? We’ve been using it for 25 years.
We have a wealth of experience in designing trading technology – from algorithm creation to automated execution and trade management. We pride ourselves on creating truly robust and quality applications – and believe we have a unique ability to bring the technical and commercial sides of the business together.
Products & Tools
Strategy Combination Tool
Looking for an easy way to estimate the combined performance of multiple trading strategies without building a single model encompassing all the rules? Our Equity Combo Utility might be just the solution you’ve been looking for. You can read about the tool and its features in this short User Guide.
For the low price of US $149, you’ll receive a package including the latest instructions, the tool itself, and a file with AFL snippets that can be inserted into your existing AFL files. You’ll also be entitled to free bug fixes and upgrades for 6 months from your date of purchase.
Need even more features? You can customize the tool yourself or give us a call to discuss additional functionality.
Testimonials
I’ve worked with Matt for about 3 years, going back to the Connors Research Chairman’s Club. When he first got my attention was two Amibroker training courses he taught. Starting from zero knowledge of Amibroker, by the end of these courses, I was empowered to be able to code and conduct portfolio-level backtests of a wide range of mean reversion strategies. This was absolutely one of the best investments in my trading skills I ever made. Wanting to go further, I followed up with Matt several times for personal consulting and he was able to help me with implementing scale-in trades through the low level CBT in portfolio level backtesting and outputting results on a monthly basis to Excel. Not only did he help me with the actual code, he also provided a detailed understanding of how it worked that enabled me to build upon it further. Overall, Matt has been a great person to work with, he is a fantastic teacher and very committed to delivering value to his clients.
We had coded a stock-trading strategy in AmiBroker (AB) using an easy high-level approach. Although our code was functional, eventually we realized that there were too many differences between the code and our actual trading, and that our code was not strictly tradeable. We had hit a wall and did not have the experience or skills to improve the code any more.
I enrolled in Matt’s Amibroker CBT Intensive course and got a lot more than I bargained for. I expected something along the lines of “do this, then do that”, etc. Instead, I received expert explanations of how the CBT works, its order breakdown, etc that I could apply across all parts. I’ve been staring at back-testing code for years and, while I could follow the logic, never understood how it all held together. This course was a big deal for me. I did the homework Matt set and applied it to one of my systems. Eureka! It worked!
I found Matt’s CBT Course to be invaluable. Having used Tradestation for many years, I began working with Amibroker about a year ago. Matt’s help was a massive shortcut for me. There are many possible pitfalls when designing and testing systems. Improper coding is certainly one of them. And with Matt’s help, I was able to learn how to code in Amibroker both accurately and efficiently. The skeleton code that comes with the course is something I’ve used over and over. And the ability to go back and review the videos and PowerPoint slides allows me to easily refresh myself on concepts I may not have used for a while. I highly recommend Matt and the CBT course.
I took Matt’s beginning and advanced Amibroker (AB) courses a few years ago. After that, I was able to code three swing-trading strategies. I then worked with Matt to put them in a live-trading form, which I’ve been trading successfully for several years on a daily basis. I also took Matt’s CBT and Adaptive Trading Course in AB in 2018. I then worked with Matt to significantly upgrade one of my 3 strategies and started trading it in Jan. 2019. Not only have my trading results more than paid for the course fees, but they have paid for several family vacations! I highly recommend Matt’s courses if you want to create trading strategies you can have confidence in.
If you have the Amibroker basics worked out, and you want to progress to the level of creating robust, accurate and incredibly powerful back-tests, Matt will guide you through it brilliantly in this course. Do this if you don’t want to stay an amateur but want the easiest possible learning curve! Matt is an excellent instructor and he really knows his stuff.
Being a trader I always look for strategies to add to my portfolio to diversify and have smooth equity curve. I worked with several Indian programmers but never reached my expectation of realistic portfolio management techniques, optimization of the system for different market condition, consultation to make good system better. Matt provided me with all of the above. His experience in coding your logics, Amibroker, backtesting and optimization is commendable. He also consult me to make sure that my systems are as realistic as possible when I trade the systems live. Initially I was sceptical that our different time zone will not lead the desired result but Matt understood the concepts(even complicated ones too) very easily and coded and provided the results at fastest turnaround time possible. He keeps on helping me with how I can bring some new filters or condition to make the system better. I can summarise his consultation fees as “Value for Money”. I advocate every new enthusiastic programme trader, experienced traders to work with him to see how he can help your system make better return.
The Only Shortcut to Success
Getting the algorithm development process right from the start is critical to success. The nuances of creating a robust trading system are many, including being aware of faulty assumptions, commissions, slippage, liquidity, survivorship bias and the list goes on. The greatest of these curses are rooted in human behavioral biases, such as natural tendencies to curve-fit and data-mine, producing wonderful backtests that utterly fail when released on an unknown future. We have a process for designing robust systems – believing there are “frameworks” in which one should operate to avoid over-optimization. This is where art meets science and why so very few ever get it right.