Today, we answer the question “How do I start algorithmic trading?”
We will specifically look at some algorithmic trading courses as well.
Contents
Introduction
Okay, so you have read our other two articles on the best algorithmic trading platform and have decided on an algorithmic trading strategy.
Now how do you start algorithm trading?
To become an algorithmic trader, you have to combine the multiple disciplines of trading, programming, and data science.
The algorithmic trader is sometimes referred to as a “quant”.
This is not a pejorative term. It is just short for “quantitative analyst”.
Such an analyst has finance, data analysis, and computer programming skills.
To get an overview of what quants do and the disciplines that they work within, the courses that Quantinsti offers is a good start.
As an example, there you will find courses such as:
Automated trading using Python and Interactive Brokers
and
Advanced Algorithmic Trading Strategies
These courses are around 50 hours long.
Hence the prices of the courses are not trivial.
Here is a tip, why don’t you just search the course instructor’s name on YouTube and see if you can find them giving various talks just to see their teaching style.
Learn Trading Concepts
In order to teach a computer how to trade, you need to become familiar with trading concepts such as:
- Bid ask spread, spillage and liquidity
- Size of position (fixed or variable position sizing)
- Risk-reward ratio
- Difference between market and limit orders
- Risk management (when to cut a trade)
- and so on
But more importantly, you need to have a profitable trading strategy with an edge.
This could be a mean-reversion strategy, for example.
Or it could be a momentum strategy or a trend-following strategy.
The rules of this strategy need to be clearly defined.
And there needs to be a rule for every conceivable situation.
You will be coding the rules of this strategy into your algorithmic trading program.
This is known as an algorithm (or algo for short).
Start out by programming one algo and running.
As mentioned in the previous article, the ideal system may involve running multiple non-correlated algos simultaneously.
But let’s walk before we start running.
Programming Courses
Next you will need to be able to code your strategy into a running computer program.
While it can be done via many different programming languages, Python is a very popular language and is well suited for writing algorithmic trading software.
The Anaconda Python Distribution is often recommended.
You may want to import Python libraries, such as pandas, numpy, and matplotlib for large data manipulations and charting functions.
If you don’t already have a favorite IDE (interactive development environment), consider Microsoft Visual Studio Code and Spyder, both of which are free.
The former is general-purpose, and the latter is more specific for Phyton users (specifically designed for scientific computing).
Udemy is the best place to look for online courses on programming languages, data science, and cloud services such as AWS and Azure.
Their courses are well structured and are of high quality.
Just watch their website every couple of weeks, and you will often find a banner displaying sales and discounts on a large number of their courses.
This often drops the price of the courses down to $15, a value that is difficult to beat.
You will get dozens of results by searching “algorithmic trading” on Udemy.
Here are just a couple of examples:
Python for Finance and Algorithmic Trading with Quantconnect
Algorithmic Options Trading on Interactive Brokers Platform
Concepts that you should look to learn are:
- How to make API calls using “post” and “get” requests
- What are REST APIs
- Programming structure: loops, arrays, maps, conditionals, object-oriented programming
- Cloud services — in particular, Amazon AWS and Microsoft Azure (pick one)
Submit To Contests
A fun way to learn is to start by developing algorithms to be submitted to Quantiacs contests.
Maybe your algo will pass their evaluation filters and win a place to be traded with live money, in which case you would receive 10% of the profits.
What?
Only 10%?
That may not sound like a lot.
When you consider that you take no downside risk, then that sounds like a good deal when you are just learning.
FAQs
Is Algorithmic Trading Profitable?
Yes, it can be. Many of the largest hedge funds are using algorithmic trading.
But you don’t have to own a hedge fund to do algorithmic trading.
It is possible for a solo retail investor to write their own program to connect to the broker’s API, and trade a profitable strategy.
This is not to say that it would be easy.
It requires a lot of hard work and dedication.
As outlined in this article, there are a lot of topics to learn first.
How much money is needed for algorithmic trading?
You need the minimum deposit amount required by your specific broker.
If you are running your algorithmic trading program on the cloud, you will have to have funds to pay for the cloud computing services.
They are on a per-usage basis.
So it will depend on how much computing power and memory your program needs to run as well as how often it runs.
Cloud computing costs have decreased significantly in recent years and are quite affordable to most individual retail investors.
We are simply talking about an algo program reading price data and executing trades for you based on that data.
We are not talking about the high-frequency algos used by institutions and hedge funds used in statistical arbitrage or anything like that.
Those require costly servers and bandwidth to achieve millisecond latency trades beyond most retail investors’ access.
Is algorithmic trading safe?
It is only as safe as the quality of the code you write.
Needless to say, back-test your strategy and code-test your program.
If you are starting out, run your algorithmic trading on a separate account with limited funds.
Don’t let its first run be on your retirement nest egg.
Do I do algorithmic trading?
No.
I don’t trust my code enough to let it trade my real money for me.
LOL.
But you may be a better programmer than I am.
I wish I had the time to learn all this stuff, though.
Conclusion
Getting started in algorithmic trading is a lot of work.
Anything worthwhile will be a lot of work.
If this topic is of enough interest to you, then it will sustain you through the long hours to come out with something worth all the effort.
We hope you enjoyed this article on how to start algorithmic trading.
If you have any questions, please send an email or leave a comment below.
Trade safe!
Disclaimer: The information above is for educational purposes only and should not be treated as investment advice. The strategy presented would not be suitable for investors who are not familiar with exchange traded options. Any readers interested in this strategy should do their own research and seek advice from a licensed financial adviser.
Interesting but perhaps I am arriving at the end of your introduction – where the material gets more difficult to understand or use.
I’ll keep looking.
thanks