Algorithmic trading is here to stay. Billions of shares still trade on the floor each day, but the majority of those buy and sell orders are done by computers.
Gone are the days of the specialist, market-maker or floor trader…. Algorithmic trading is a process that uses computers, to place trades perfectly. The key benefit is the computer and the algorithm, never breaks your rules.
This method is often called algo trading. Other variations include automated trading, and black-box trading. To give you a complete picture, we should also mention gray-box trading. A gray-box allows for discretionary decisions by the trader. Algo trading is fascinating and mysterious, but it simply means your trade ideas, are executed flawlessly. The computer does all the work, after you input your criteria. Step 1 is crucial to the process.
A well defined edge, identifies the opportunity. This algorithm is pure. There are no qualifiers to fine-tune the edge. Developing an edge, and converting it into programming code, is where the money is earned in algorithmic trading. Qualifiers force price action and volume, to unfold according to our plan, or we do not enter a new trade.
There has never been a better time to become an algo trader or developer. A winning algo edge, means you have identified a moment in price, volume and time, that occurs more often than not. You are seeking a reason to allocate capital, because you believe the potential profit, is worth the potential risk. Algorithmic trading strategies and programs, scan all available data, and execute trades when your edge is valid.
Identifying an edge is rather simple. Choosing the best qualifiers that match your goals, resources, and capital is where your algo becomes special. There are essentially three best-practices to validate your algo strategy: Back-testing an algo strategy involves simulating the performance of a trading strategy using historical data. This means you test a strategy, using price action that has already occurred. This form of validation, gives you an opportunity to estimate the effectiveness of your edge.
It should not be used as final validation, but works well to determine if your edge is worth pursuing. One caveat to consider with back-testing, and then analyzing your results, is the trap of optimization. This is a vicious trap of perfection. Once you have preliminary validation, move onto simulated trading. Simulated trading, tracks your algo strategy against live market data. You get results and feedback without the benefit of knowing the outcome of price action.
In essence, you cannot choose the perfect day to validate your edge. This process is obviously slower, because you can only test one day at a time. The benefit is you cannot make tweaks in hindsight. You let your algo strategy run the entire day and then review the data for any possible changes. Live trading to validate your algo strategy is by far the most effective method for a true validation. You get feedback that shows actual executions, and how your trading program performed within the two critical market conditions of, liquidity and volatility.
While valuable, back-testing and simulated trading provide feedback for trades that never occur. This can give false hope. Because back-testing and simulated trading never add or removes shares from a market, you will truly never know performance until you attempt trades that interact with available shares in the market.
Liquidity identifies the ease with which you can execute a trade, because there are shares quoted at the bid or ask, and your algo, and a transaction took place. As you develop and test your algorithmic strategy, you must factor in the contract size or share size you plan to trade, and the ease with which you can reasonably execute that trade. Slippage means you anticipate not receiving the perfect fill price that you received while back-testing or simulated trading.
Large orders, without liquidity, can be a slippage disaster. Volatility represents, how fast and how far, a security moves, within a designated period of time. In trading lingo, many who use technical analysis determine volatility, by using the Average True Range indicator. ATR determines how far a security trades from high, to low over a designated period of time. This means if you are trading AMZN, the swings are much wider and share size must match your risk tolerance.
The same applies to futures contracts. Liquidity and volatility are key elements to consider when validating your algo.
There are literally thousands of potential algorithmic trading strategies, here are few of the most common to jump start your journey:. Your edge is determined by identifying an obvious direction to order flow. This edge could be over months, or over minutes. The key to success with this strategy is defining the time frame to operate.
The objective is to pick a side, then pick a spot to enter. The shorter the time frame, the more frequently you will trade because the trend will change quicker and you will receive more signals. Momentum algos look for the futures contract to move quickly in one direction on high volume. This edge seeks to quickly enter on a pause, ride the momentum, and then exit on the next pause.
This algo does not ride big winners. The plus side is it should not have big losers either. Momentum strategies in the direction of the order flow, are generally regarded as smart trading. This last statement is especially true because of algorithms! There was a period in time, when price action had a nice fluid back-and-forth rhythm.
Algos have changes that dramatically. Leaving no reprieve for the counter-trend neophyte. Reversion to the Mean Algo Strategies: This is reversion to the mean algo trading. The goal of this trade, is to time the entry, at an extreme price point, anticipating a profitable reversal.
Certain markets, offer opportunities to track large buyers and sellers. Tighter spreads and faster computers, have made this challenging for the manual trader. One door closes and one door opens, scalping opportunities have opened for smart algo developers and traders.
This is the algo that gets all the publicity. The perceived money-machine for the privileged quant-wizards. The ever expanding industry of computerized trading, is a changing landscape that appears to have no bounds, save imagination, and computing speed. For all the fancy trader lingo, this is simply automated trading. Visual programming language, allows futures and options traders to design, create and deploy automated high frequency trading algorithms without having to write a single line of code.
With an easy-to-use, drag-and-drop interface, users apply building blocks to construct circuit-like designs on their computer screens. The language and program, offers the flexibility to design your own strategy, and the opportunity to study and implement, pre-made strategies. ADL makes algorithm design accessible to anyone, not just advanced programmers.
ADL provides safety measures at design time and at run time that are not available in traditional programming context, thereby reducing risk and the time required to design, create and test programs while providing a safer trading environment. Java is popular and with good reason.
It can be debugged, which places an emphasis on checking for errors. Python is known as, an object-oriented language. The programming language is interactive, and portable, which makes it easy to work with for professional coders.
This general-purpose language is typically used in systems programming, and is quite popular. It was designed with a bias toward system programming and embedded, resource-constrained and large systems, with performance, efficiency and flexibility of use as its design highlights. What is Algorithmic Trading? The algo, is a set of specific criteria, that: Finds trades that match our edge.
Identifies the predefined entry criteria. Place the trade entry. Analyzes and tracks price movement, bids, offers and transactions. Identifies the predefined exit criteria. Places the exit orders to complete the trade. B Sell the new position, any time the price has a. How to Develop a Profitable Algorithmic Strategy A winning algo edge, means you have identified a moment in price, volume and time, that occurs more often than not.
The trading term for this is trade expectation.More...