Ultimate List of Automated Trading Strategies You Should Know - Part 1

This is the part 1 of a series “Ultimate List of Automated Trading Strategies

Photo by  Artem Bali  on  Unsplash

Photo by Artem Bali on Unsplash

So many types of automated trading use-cases

Since the public release of Alpaca’s commission-free trading API, many developers and tech-savvy people have joined our community slack to discuss various aspects of automated trading. We are excited to see many have already started running algorithms in production, while others are testing their algorithms with our paper trading feature, which allows users to play with our API in a real-time simulation environment.

When we started thinking about a trading API service earlier this year, we were looking at only a small segment of algo trading. However, the more users we talked with, the more we realized there are many use cases for automated trading, particularly when considering different time horizons, tools, and objectives.

Today, as a celebration of our public launch and as a welcome message to our new users, we would like to highlight various automated trading strategies to provide you with ideas and opportunities you can explore for your own needs.

Please note that some concepts overlap with others, and not every item necessarily talks about a specific strategy per se, and some of the strategies may not be applicable to the current Alpaca offering.

(1) Time-Series Momentum/Mean Reversion

Background

(Time-series) momentum and mean reversion are two of the most well known and well-researched concepts in trading. Billions of dollars are put to work by CTAs employing these concepts to produce alpha and create diversified return streams.

What It Is

The fundamental idea of time-series forecasting is to predict future values based on previously observed values. Time-series momentum, also known as trend-following, seeks to generate excess returns through an expectation that the future price return of an asset will be in the same direction as that asset’s return over some lookback period.

Trend-following strategies might define and look for specific price actions, such as range breakouts, volatility jumps, and volume profile skews, or attempt to define a trend based on a moving average that smooths past price movements. One of the simple, well-known strategies is the “simple moving average crossover”, which buys a stock if its short-period moving average value surpasses its long-period moving average value, and sells if the inverse event happens.

Mean-reversion is the expectation that the future price return of an asset will be in the opposite direction of that asset’s return over some lookback period. One of the most popular indicators is the Relative Strength Index, or RSI, which measures the speed and change of price movements using a scale of 0 to 100. For the purposes of trying to assess the likelihood of mean-reversion, a higher RSI value is said to indicate an overbought asset while a lower RSI value is said to indicate an oversold asset.

For Implementation

Trend-following and mean-reversion strategies are easy to understand since they look at a single asset’s time-series and try to make a prediction about that asset’s future return, but there are many ways to interpret the past behavior. You will need access to historical price data and may benefit from an indicator calculator library such as TA-lib. Virtually every trading framework library, including pyalgotrade, backtrader, and pylivetrader, can support these types of strategies.

Here is the Quantopian tutorial with backtest result for moving average crossover: Quantopian Tutorials

(2) Cross-Sectional Momentum/Mean Reversion

Background

In the U.S. stock market, there are more than 6,000 names listed on the exchanges and actively traded every day. One of the hardest problems in stock trading (and also true for global cryptocurrency trading) is how to pick the stocks.

What It Is

Cross-sectional momentum compares the momentum metrics across different stocks to try to predict the future returns of one or more of them. Even if two stocks such as Facebook and Google are indicating a momentum breakout, this may be driven by the market, but you try to beat the market by taking stronger momentum between those signals. Same for mean reversion. The point is that we consider the market movement that drives each individual stock and consider the relative strength of signals across stocks in an effort to produce a strategy that will outperform the market. This tends to be more computationally heavy, since you need to calculate the metrics with potentially tens to hundreds of time-series.

For Implementation

Again, for this type of strategy libraries like TA-Lib may make it easier to calculate the indicators. Also, you may need simultaneous access to multiple symbols’ price data. IEX’s API can provide up daily bar data for up to 100 stocks per query.

A medium post about cross-sectional study: Basics of Backtest and Cross-sectional Momentum

(3) Dollar Cost Averaging

Background

This is one of the simplest automated trading strategies and it is widely used by many investors.

What It Is

The idea is to invest a fixed amount of money into an asset periodically. You may doubt it, but some research indicates that this works in the real world, especially long-term. The logic behind it is that price fluctuates many times, and you may buy the stock cheaper overall compared to just investing in the stock at one point in time.

Remember, all of you who contribute to your 401k account are basically doing this. However, you might never think about doing it yourself, simply because there has been no easy way to automate this process.

For Implementation

Now with Alpaca trading API, it’s much simpler and provides much more flexibility.

(4) Market Making

Background

Market makers are important intermediaries who stand ready to buy and sell securities continuously. By doing this, they provide much-needed liquidity and are compensated for their inventory risk primarily by capturing bid-ask spreads.

Market making used to be done primarily by humans, who worked as floor traders in the pits, but now it’s almost entirely performed by machines. As exchanges have become more and more electronic, the strategy market makers employ has naturally required automation.

What It Is

There are a variety of approaches to market making but most typically rely upon successful inventory management through hedging and limiting adverse selection.

Some market makers may have very tight exposure limits and seek to turn over their positions quickly with the goal of being flat at the end of each day. Others may operate on a much longer horizon, carrying a large and diverse portfolio of securities long and short indefinitely. Undoubtedly, for any market maker, speed helps. The speed of calculation allows the market maker to continuously update its pricing and portfolio risk models, while the speed of execution allows the market maker to act on its models in a timely manner in an effort to reduce adverse selection and get better pricing on its hedges.

Competitive market makers need high-resolution data and a low latency infrastructure, although typically the longer their trading horizon is, the less sensitive they are to these things, and a smart but slow model goes a long way.

For Implementation

Also, in order to process vast amounts of data quickly and handle concurrency, languages like python may not be suitable. Go/Rust would be a good choice for balance between ease of concurrency handling and processing speed, as well as functional languages like Erlang/OCaml or good old languages like C++.

Some high-level explanation of market making: How profitable is market making on different exchanges

(5) Day Trading Automation

Background

Lots of day traders develop their trading strategies based on a mechanical set of conditions that are first based on intuition. Since manual day trading involves continuously assessing market conditions and making discretionary trading decisions on the spot, it can often be very physically and emotionally draining. Because the strategies are based on some rules or heuristics which can be codified, it is natural to think they can be automated, which is likely the case.

What It Is

One of the very well-known day trading strategies is the gap-up momentum strategy.

Suppose between the previous market close and next market open there is a positive earnings report. The market opens with a big gap, drawing lots of traders’ attention, and the price keeps going up for a while in the morning (but may not continue for long).

This strategy seeks to capture this follow-through momentum. The challenge here is that not all gap-up stocks keep going up, and among a handful of screened stocks, you need to watch each one’s price action simultaneously.

Some traders may enter on a price breakout from a certain price resistance level, while others may wait to see a chart pattern form to determine the first bottom before going higher. Day trading often relies on analyzing the stock’s price chart and fine-tuning the algorithm to capture the price action can be tricky. That said, once it’s well developed, you are letting your bot trade on your behalf as if you were trading manually, and now you don’t need to monitor the markets and you can also monitor more stocks at the same time without any emotions affecting your trade execution, which is very compelling.

For Implementation

The main thing you need for this is access to market data. You may not even need indicator calculations but instead, you may need a stock screening library such as pipeline-live. The latency typically isn’t so important, so you don’t need to write your system in C++. Python, as well as other lightweight languages, are likely sufficient.

Some reference: Momentum Day Trading Strategies for Beginners: A Step by Step Guide

To Be Continued…

This is part 1 of 3 posts to overview the various types of automated trading strategies. Stay tuned for our next post to cover more.

/

Algo Trading News Headlines 9/18/2018

5 Reasons Why Cryptocurrency Trading Bots Are So Popular

(www.cryptodisrupt.com)

“Have you been considering using cryptocurrency trading bots or are looking for a way to get involved with crypto trading on exchanges? Many people are thinking the same as you. Here are five reasons why trading bots are so popular.”

Photo by  Fancycrave  on  Unsplash

Photo by Fancycrave on Unsplash

Why Robots Are Bad Financial Advisors

(www.nasdaq.com)

“In an age of rapidly advancing technology, more investors are opting for DIY financial planning and investment management platforms. These trading platforms, retirement calculators, and auto-rebalancers are increasingly sophisticated, but many investors will learn in the next major market meltdown that there is a human element that cannot be replicated by even the most advanced of these tools.”

Can Computers Time The Market?

(www.seekingalpha.com)

“With my background in software, I decided to design, develop, and test out various quantitative models, and leveraged AI to determine the optimal values for each model or “strategy.” Though some excellent quant platforms like Quantopian exist, I opted to develop this simulator from the bottom up.”

10 Years Later, Many Deep Scars From the Financial Crisis Remain

(www.247wallst.com)

“24/7 Wall St. has evaluated those scars. Admittedly, this has more of an American focus than an international focus. Also worth noting is that an indefinite number of additional scars will remain in place for years and are the aftermath of the Great Recession and financial crisis of the past decade. There is also no way to categorize which scar is the most prevalent because that varies from person to person and from group to group.”

/

Algo Trading News Headlines 9/17/2018

The who, how and why of high-frequency crypto trading

(www.bravenewcoin.com)

According to the Financial Times, several leading high-frequency trading houses, including DRW, Jump Trading, DV Trading, and Hehmeyer Trading have entered the crypto asset markets last year. Several newly-launched crypto hedge funds are also using algorithmic trading strategies to generate a return on investment for their investors.

Dutch high-frequency trading house, Flow Traders BV, also recently made a move into the crypto markets, according to Bloomberg. The Amsterdam-based company is making markets in exchange-traded notes linked to bitcoin and ether due to strong investor demand for crypto investments.

Photo by  Marc Szeglat  on  Unsplash

JP Morgan Says Severe Crisis to Arrive in 2020

(www.nasdaq.com)

The consensus is that there will be a major “liquidity crisis” with huge selloffs in major asset classes, and no one to step in to buy. The losses will be exacerbated by the shift to passive management and the rise of algorithmic trading. JP Morgan says that the Fed and other central banks may even need to directly buy stocks, and there could even be negative income taxes. The bank thinks the crisis will hit sometime after the first half of 2019, most likely in 2020.

Billionaire who once built robots to trade goes to war with them

(www.economictimes.com)

Thomas Peterffy helped launch the electronic-trading revolution that transformed the US stock market. And while the billionaire hasn’t soured on automation, he’s taking a lead role fighting back against the speediest traders. 

Interactive Brokers Group Inc. announced Wednesday that it will list its shares on an exchange run by IEXNSE 0.06 %Group Inc., which was made famous by Michael Lewis in “Flash Boys.” The 2014 book documented the market’s efforts to use a 350-microsecond speed bump to eliminate advantages IEX believed the fastest traders had in US stocks. When shares of Interactive Brokers move over from Nasdaq Inc., it will be IEX’s first win in its delayed plan to list corporations.

Quant Strategy in Emerging-Market FX Posts Best Run in Six Years

(www.bloomberg.com)

A Nomura index that mimics a trend-following strategy by chasing momentum in 10 EM currencies against the dollar has outperformed the JPMorgan Emerging Market Index by nearly 20 percentage points so far this year.

/

Algo Trading News Headlines 8/29/2018

Brazilian crypto trading platform hacked, over 264000 user data leaked

(www.coingeek.com)

Portal do Bitcoin reported that the hack was initially brought to light through a YouTube video posted by Investimentos Digitais. A total of 14,500 accounts containing a total of 5,813 BTC, currently worth about $40 million, were reportedly affected by the security breach.

Photo by  Pedro Menezes  on  Unsplash

Photo by Pedro Menezes on Unsplash

Forecasting Market Movements Using Tensorflow

(alpaca.markets)

In this post we’ll be looking at a simple model using Tensorflow to create a framework for testing and development, along with some preliminary results and suggested improvements.

$6 Billion Daily Crypto Volume is Being Faked, How Can it be Combated?

(www.ccn.com)

Earlier this week, the Blockchain Transparency Institute (BTI) published a report claiming that the global crypto exchange market is faking $6 billion of its daily volume. The researchers at BTI evaluated the user activity and traffic of the market’s biggest crypto exchanges, comparing their projected trading volume to other metrics.

Empower Sebi to crack down on erring CAs, says panel

(http://www.asianage.com/)

New Delhi: The fair market committee report submitted to securities market regulator Sebi has many more unpleasant surprises for various market-connected entities. For instance, it opens a pandora’s box on whether Sebi has jurisdiction over errant chartered accountants or not.

/

Algo Trading News Headlines 8/27/2018

Trading Places — Lines Blurred Between Traders And Programmers

(seekingalpha.com)

The recent WSJ article focused upon details from Adam Korn, a 16-year veteran at Goldman. He stated that success today depends less on trusting one’s gut, rather much of a trader’s job is embedded in the computer code or algorithms, which do much of the work now.

What is the real story though, what has all this computerized algorithmic trading truly done, how much value has it truly created? One question I would like to ask, is there a correlation between the explosion of our debt levels and this newly digitized financial age?

Photo by  Phil Botha  on  Unsplash

Photo by Phil Botha on Unsplash

Major Russian Airline Tests Blockchain in Bid to Track Fuel Payments

(www.coindesk.com)

According to S7, the application shares data about fuel demand on a shared ledger, a copy of which is managed by each of the three parties. Further, payments for the fuel can be conducted on the network, with digital invoices created via smart contract during each transaction.

Python Notebook Research to Replicate ETF Using Free Data

(alpaca.markets)

ETF is one of the great investment products in the last decade, and it has allowed so many people to gain the exposure to the wide range of assets easily at low cost. It is easy to buy a share of ETF without knowing what’s in there, but as a tech-savvy guy yourself, you may wonder how it works. By reconstructing the fund yourself, you may even come up with something better.

Trading Lesson: Don’t Touch That Dial. More to Come, Hedge Your Bets

(www.moneyshow.com)

In my 30 years as a trader, I’ve never seen a market like this. If the bots remain faithful to their programs, we are still likely to see higher stock prices over the next few weeks to months.

STEROID Launches New Automated Cryptocurrency Trading Algorithm

(www.bitcoinexchangeguide.com)

Algorithms run our online world, for the most part, a majority of everything done online is associated with an algorithm in one way or another. It only makes sense therefore that they would be used in the financial world as well. That is why STEROID has been developed, to create a functioning opportunity for traders on crypto exchanges.

Here’s how artificial intelligence can be used to beat the market

(www.cnbc.com)

CNBC’s Bob Pisani is joined by Sam Masucci, ETF Managers Group CEO, to discuss how he’s using an AI program to pick stocks.

/

Algo Trading News Headlines 8/23/2018

Barclays hires AI specialist from Goldman intrading revamp

(www.fnlondon.com)

The hire is another sign of Barclays’ intent to ensure its markets business can compete with rivals in the City and on Wall Street, which are increasingly turning to new technologies to improve performance and cut costs.

No Quant Is Safe as Global Stress Hits Risk Parity Where It Hurts

(www.bloomberg.com)

Risk-parity’s travails might test faith in the trading style over traditional allocations. It’s among the worst-performing categories tracked by JPMorgan, falling 1.2 percent this year. That compares to a 2.1 percent gain for long-short equity funds, 1.7 percent for macro funds, and 2.9 percent for balanced mutual accounts, according to data published by the U.S. bank earlier this month.

UBS trials Netflix-style algorithms for trading suggestions

(www.cnbc.com)

UBS is looking at applying recommendation algorithms to suggest trades to its asset management and hedge fund clients, similar to those used by a host of consumer technology companies.

Market makers, takers and fakers: US exchanges are losing fast

(bravenewcoin.com)

The great ETF debate rages on and there’s a lot of conjecture as to why the SEC should or shouldn’t approve one. But what needs to be addressed foremost is the immaturity and unreliability of many of the exchanges that bitcoin derives its “global spot price” from — that number now over 210 — and the obstacles they may encounter going forward.

Indicators That Could Have Made You 626% ROI During The Recent Bloodbath

(cryptodaily.co.uk)

Those indicators are the 5 EMA and 10 EMA as can be seen on the daily chart for BTC/USD. These indicators have been signaling the direction of every bullish and bearish wave throughout the correction period.

/

Algo Trading News Headlines 8/17/2018

Startup Raises $23 Million To Make Crypto Trades Faster And Stealthier

(www.forbes.com)

Silicon Valley-based cryptocurrency trading platform Sfox has raised $23 million in new investment, led by venture firms Tribe Capital and Social Capital. The 20-person startup aims to help investors make large trades by routing their orders to multiple places, enabling faster execution and better prices. Khosla Ventures, startup accelerator Y Combinator, and crypto investors Blockchain Capital and Digital Currency Group also participated in the Series A funding round. Forbes estimates Sfox has reached $15 million in revenue over the past 12 months.

Photo by  John-Paul Henry  on  Unsplash

Blockchain Firm Exscudo To Add Trading Robots Support On Its Crypto Exchange

(blocktribune.com)

Exscudo is currently testing a trading terminal API that will allow Exscudo exchange users to utilize a trading robot that automatically makes trading decisions based on market data analysis and pre-programmed set of parameters.

Forecasting Market Movements Using Tensorflow — Intro into Machine Learning for Finance

(blog.alpaca.markets)

In this post we’ll be looking at a simple model using Tensorflow to create a framework for testing and development, along with some preliminary results and suggested improvements.

The World Economic Forum warns that AI may destabilize the financial system

(www.technologyreview.com)

Compiled through interviews with dozens of leading financial experts and industry leaders, the report concludes that artificial intelligence will disrupt the industry by allowing early adopters to outmaneuver competitors. It also suggests that the technology will create more convenient products for consumers, such as sophisticated tools for managing personal finances and investments.

/

Algo Trading News Headlines 8/15/2018

Goldman Sachs, JPMorgan Strengthen Algo Trading Units With New Hires

(www.financemagnates.com)

Algo trading has been an important component for many banks in recent months, with the latest hires possibly portending a drying up of supply to meet a growing demand. In any scenario, banks have been quickly scooping up top-tier talent in the algo trading arena, with Goldman Sachs and JPMorgan emerging as the latest players to do so.

Photo by  Sean Pollock  on  Unsplash

Photo by Sean Pollock on Unsplash

Comparing 3 Different Types of Neural Network Architectures in Finance

(blog.alpaca.markets)

One good use may be in the analysis of portfolios. By clustering equities and financial instruments you can get a unique view of the distribution of exposure and risk, and either hedge accordingly or look to maximise the efficiency of exposure to one area of the market.

4 Problems with Using a Crypto Trading Bot on Exchanges

(cryptodisrupt.com)

“Get a crypto trading bot”, said a friend. “You will be swimming in cash”, he said. It was with the ultimate of best intentions that my friend advised me to use a trading bot. They are definitely a great way for crypto trading newbies to start making money and understanding how exchanges and the markets work but are trading bots simple to use? And is it really that easy to make money with them?

Can Robots Replace Day Traders on Wall Street?

(www.financemagnates.com)

Goldman Sachs has already begun to automate currency trading, and has found consistently that four traders can be replaced by one computer engineer.

/

Algo Trading News Headlines 8/10/2018

Is making money from forex an easy or a difficult thing?

(www.thesouthafrican.com)

Automated trading helps to put a trading plan into action without any participation, with the help of technologies. Automated trading minimizes the influence of your emotions. The advantage of this is that a trader doesn’t need to go deep into strategies and platform trading features.

Photo by  Thought Catalog  on  Unsplash

Top 5 Bitcoin Trading Bots for 2018

(nulltx.com)

When it comes to cryptocurrency trading, a lot of users prefer an automated approach. Although there are a few trading platforms which offer built-in algorithmic trading, trading bots are also very popular. It is interesting to see how the landscape has evolved in this regard. We ranked these bots based on their customization options.

Forecasting Market Movements Using Tensorflow

(blog.alpaca.markets)

Is it possible to create a neural network for predicting daily market movements from a set of standard trading indicators? In this post we’ll be looking at a simple model using Tensorflow to create a framework for testing and development, along with some preliminary results and suggested improvements.

New report examines the automated trading market forecast to 2025

(www.whatech.com)

This report studies the global Automated Trading market size, industry status and forecast, competition landscape and growth opportunity. This research report categorizes the global Automated Trading market by companies, region, type and end-use industry.

/

Algo Trading News Headlines 8/2/2018

Fondex simplifying copy and algorithmic trading for its growing customer base

(www.worldfinance.com)

Traders can create algorithms with varying degrees of complexity. They can automate only a part of their trading strategy so as to avoid monitoring a chart, or they can create complex algorithms that read multiple markets and time frames, use multiple indicators, and employ sophisticated risk management systems that even take into account calendar events and news.

Photo by  Kelly Sikkema  on  Unsplash

Former High-Speed Trading Executives Allege ‘Tyrannical Coup’ at Quantlab

(www.wsj.com)

A leadership fight has broken out over one of the world’s most secretive and profitable high-frequency trading firms, pitting a beret-wearing mathematician against a former business partner and a Ukrainian physicist.

Four simple steps to creating a Trading Bot without any Coding skills

(smartereum.com)

Most important is the fact that any trader can transform their own peculiar trading strategy into an automated system without any special expertise that is beyond typing in plain English.

Algo Trading for Dummies — Building a Custom Back-tester

(blog.alpaca.markets)

Building a back-tester is a fantastic conceptual exercise. Not only does this give you a deeper insight into orders and their interaction with the market, but it can also provide the framework for the order handling module of your trading bot.

Tradingene, the new way to invest in trading algorithms

(www.globalbankingandfinance.com)

The platform’s marketplace for trading algorithms and subscription plans were released this week, bringing trading algorithms to retail investors and creating new investment opportunities.

/