Neural network high frequency trading strategy
Although I do not exclude a future buyout, I am presently focusing on improving the product and trying to scale it. One of the things that I plan on doing soon is increasing the capital and therefore putting the bot through more trading volume. There are tons of improvements I have in mind, especially on adjusting the position-holding time span, as well as solutions to make it more lightweight, facilitating larger volumes.
I wasted way too much time trying to apply high frequency trading in Bitcoin. At first the idea sounded great, but I was soon facing a lot of technical issues trying to scale the amount of requests. However, I am not yet convinced that it's impossible to achieve true HFT with cryptocurrencies, so it might be something I come back to in the future. After drifting away from the idea of HFT due to the technical limitations, I looked into a more analytical approach in automated trading.
Most of those concepts couldn't be applied in the Bitcoin market, as it's highly unpredictable, making it hard to shape the models around it. That's when I decided to stick to the stock market. Another big mistake in the beginning was relying too heavily on models. Instead of trying different approaches in analyzing the data I had, I relied solely on the models for identifying profitable patterns without investing time into other more direct solutions. Models are only simple real world abstractions, and my common sense has saved me more than once.
Now this is not by any means a reliable metric, and there are many factors that affect it. The bot has not been tested enough to guarantee that this isn't just a fluke it might as well be. Large investment management companies would do anything to achieve those statistics, and I'm sure I won't keep up that amount of success in upcoming trades. The success so far was also greatly impacted by the favorable market conditions, chosen stocks, and the fact that the bot was running intermittently.
I learned this the painful way. Not too long ago the market went pretty crazy, and I'd be lying if I said that I wasn't expecting some major crashes of the stocks I was trading. Although my stop-loss saved me from some brutal losses, had I not stepped in at the right time, the bot would've ruined all the profit from the past months. That event really got me thinking, and I decided to stop it running for a few days until I fixed that loophole.
This was also a great learning experience for me, and I believe that without going through those ups and downs, I would've never managed to get the algorithm to where it is today. I have no regrets losing time on Bitcoin, as it gave me a deeper understanding of how cryptocurrency trading works, which might prove useful some day. Probably my biggest single advantage is being a starry-eyed young dreamer.
To some extent, this allows me to believe enough to put effort into ideas in that others wouldn't. That's what motivated me to persevere in finding those "backdoors" in the market. While many people believe individual traders don't stand much of a chance against the well-equipped companies, I am here to prove that with the right implementation there still is plenty of space in the market.
Another immensely helpful resource were the public research papers available online. In fact, I got tremendous help from papers published back in I often found that most of them are easily overlooked, although they contain super useful analyses. Being a workaholic has also contributed a fair amount to this success. I have no issue whatsoever working hours per day. With time, I developed a very productive and consistent lifestyle, managing to get rid of most distractions.
This allowed me time to invest in polishing and researching the different strategies for this project. If you've worked your butt off to build something and give up on launching it, no one will care about it. We live in a very capitalist society where people will judge you based on real results.
No one cares about your initiative and the reasons why you didn't launch. As Sam Altman says, nothing will excuse you for not having a great product.
However, not having anything is certainly worse than that. Don't make it perfect from the first version. Test the market first, gather tons of feedback and constantly iterate over your idea. Although this is not necessarily a customer-focused product yet?
Side projects allow you to experiment on crazy ideas without being labeled as crazy. And definitely go for the craziest idea you have in mind. That's how most of the successful companies started talk Facebook, Uber, AirBnb. Every problem has a solution. You just have to be creative enough to find it.
I am currently available for freelance work. You should join the Indie Hackers community! We're a few thousand founders helping each other build profitable businesses and side projects. Come share what you're working on and get feedback from your peers. Not ready to get started on your product yet? The community is a great place to meet people, learn, and get your feet wet.
Feel free to just browse! Take it into the bank? Give it to a hedge funds? Which gives us a slightly bad conscience , since those options are widely understood as a scheme to separate naive traders from their money. And their brokers make indeed no good impression at first look. Some are regulated in Cyprus under a fake address, others are not regulated at all.
They spread fabricated stories about huge profits with robots or EAs. They are said to manipulate their price curves for preventing you from winning. And if you still do, some refuse to pay out , and eventually disappear without a trace but with your money.
Are binary options nothing but scam? Or do they offer a hidden opportunity that even their brokers are often not aware of? Deep Blue was the first computer that won a chess world championship. That was , and it took 20 years until another program, AlphaGo , could defeat the best human Go player.
Deep Blue was a model based system with hardwired chess rules. AlphaGo is a data-mining system, a deep neural network trained with thousands of Go games. Not improved hardware, but a breakthrough in software was essential for the step from beating top Chess players to beating top Go players. This method does not care about market mechanisms.
It just scans price curves or other data sources for predictive patterns. In fact the most popular — and surprisingly profitable — data mining method works without any fancy neural networks or support vector machines. This is the third part of the Build Better Strategies series. As almost anything, you can do trading strategies in at least two different ways: We begin with the ideal development process , broken down to 10 steps.
We all need some broker connection for the algorithm to receive price quotes and place trades. Seemingly a simple task.
Trading systems come in two flavors: This article deals with model based strategies. Even when the basic algorithms are not complex, properly developing them has its difficulties and pitfalls otherwise anyone would be doing it. A significant market inefficiency gives a system only a relatively small edge. Any little mistake can turn a winning strategy into a losing one. And you will not necessarily notice this in the backtest.
The more data you use for testing or training your strategy, the less bias will affect the test result and the more accurate will be the training. Even shorter when you must put aside some part for out-of-sample tests. Extending the test or training period far into the past is not always a solution. The markets of the s or s were very different from today, so their price data can cause misleading results. But there is little information about how to get to such a system in the first place.
The described strategies often seem to have appeared out of thin air. Does a trading system require some sort of epiphany? Or is there a systematic approach to developing it? The first part deals with the two main methods of strategy development, with market hypotheses and with a Swiss Franc case study. All tests produced impressive results. So you started it live.
Situations are all too familiar to any algo trader. Carry on in cold blood, or pull the brakes in panic? Several reasons can cause a strategy to lose money right from the start. It can be already expired since the market inefficiency disappeared. Or the system is worthless and the test falsified by some bias that survived all reality checks. In this article I propose an algorithm for deciding very early whether or not to abandon a system in such a situation.
You already have an idea to be converted to an algorithm. You do not know to read or write code. So you hire a contract coder.
Just start the script and wait for the money to roll in. Clients often ask for strategies that trade on very short time frames.
Others have heard of High Frequency Trading: