The process in which it takes to create quantitative trading strategies is much more complete than you may think. But how they operate is actually very simple. Each trade generated are triggered by a setup of rule within the quantitative trading strategy you have built, and then the trade is sent electronically your broker to be executed. Although the mechanism for execution can be manual or semi-manual, it is advisable to automate your quant trading system as much as you can. The idea is to remain free to work with two or more strategies simultaneously, as you gradually gain expertise in quantitative trading.
There are some key things you must keep in mind when you are in the process of building a quant trading system. These include the kind of interface you are having with the broker, the way in which you can minimise your transaction costs and finally to see how divergent the performance of your strategy is in the live scenario as compared to the time it was being back tested.
Three Steps to Building Quantitative Trading Strategies
First, let’s deal with the aspect of interfacing your quantitative trading strategy with your brokerage. For a low frequency trade or LFT that is simple, platforms such as Tradestation or MATLAB are good enough, but if you have a high frequency trade in mind, it is necessary to have an execution system that is written in proficient software language such as C++. If you want to find a job as a trader in fund that uses quant strategies, you must have good programming skills, for it is often the traders who have to double up as the executors of a strategy.
The second issue, as we mentioned earlier, is keeping a check over the transaction costs. Let us tell you about the major components of the transaction costs that will help you determine how to keep a check over them. The first is the fees or the commission that will be charged by firstly the brokerage, then by the exchange and lastly the regulatory body. The next aspect of cost is covered by the slippage factor. This is the difference you had intended for your order and what it actually works out to be. Next, is the spread. This is the bid or ask price of the traded security. This does not remain constant as it is dependent on the liquidity situation in the market.
It is quite a challenge to determine the transaction costs from a backtest and your transaction costs can indeed play a big role in determining whether or not your strategy remains profitable. In bigger fund houses, there are teams that are dedicated for this very purpose of execution, but if you are on the team of a smaller fund house or are a retail trader, you need to be careful with the data you are obtaining. Alternatively, you need to be aware of hedging strategies that will protect you against the inefficiencies in your data.
The final issue that may crop in the executions of your quantitative trading strategies is the concern over the divergence of the performance of your strategy. Even if everything is right in your backtest, there may be hidden bugs that create havoc when things go live. There can also be other unforeseen circumstances such as a big change in the regulatory regime or a macroeconomic crisis that may not augur well for your strategy and keep it from becoming profitable.
While it is imperative for you to do all you can to be perfect in the execution of your quant strategy, there is no need to be crestfallen if it does not work all at once! It’s perfectly alright if your strategy does not get executed as you had perceived. Being aware of the risks as we mentioned earlier, will certainly help you cope and better your execution strategy and be back with a bang!
Have My Automated Quantitative Trading Strategies Working For You!
Chris Vermeulen
AlgoTrades.net
Auto Amazon Links: No products found. (1 items filtered out)