A rising tide lifts all boats. "Price move predictions" sounds as effective as tea leaf reading.
How I made $500k with machine learning and high frequency trading
191–200 of 320 posts
Re: How I made $500k with machine learning and high frequency trading
#192This is just glorified gambling. I am not sure what special insight or advantage he had, other than his own model. Every trader has a model. It could have easily been called "how i lost 500k with machine learning". Like gambling, it's easy to manipulate statistics to show that you did well in some period of time. I worked for a large investment bank about 10 years ago, writing trading programs for quant traders who w…
I once worked for a software shop, and part of my job was writing trading code in a proprietary language for customers, who ranged from low end day traders to 8 figure annual revenue hedge funds. I had access to all kinds of tools, and saw many a varied strategy. There's quite a lot of money to be made selling solutions. I won't day trade.
What kind of solutions?
Re: How I made $500k with machine learning and high frequency trading
#193I run a 12 person HFT group in Denver. This is an excellent description of how an individual can bootstrap themselves into success. Great story, and nicely presented. The one interesting point that he glossed over is what his indicators were. He wrote, "The indicators that were most useful were all relatively simple and were based on recent events in the market I was trading as well as the markets of correlated secur…
How well are you guys doing? How much money did you make?
Re: How I made $500k with machine learning and high frequency trading
#194I run a 12 person HFT group in Denver. This is an excellent description of how an individual can bootstrap themselves into success. Great story, and nicely presented. The one interesting point that he glossed over is what his indicators were. He wrote, "The indicators that were most useful were all relatively simple and were based on recent events in the market I was trading as well as the markets of correlated secur…
Thank you. I'm looking back through my code and there are really a lot of indicators. It would get pretty technical to explain them. They are all explainable it's just that each one corresponds to slightly different market conditions and I just didn't want to get into it.
Re: How I made $500k with machine learning and high frequency trading
#195I run a 12 person HFT group in Denver. This is an excellent description of how an individual can bootstrap themselves into success. Great story, and nicely presented. The one interesting point that he glossed over is what his indicators were. He wrote, "The indicators that were most useful were all relatively simple and were based on recent events in the market I was trading as well as the markets of correlated secur…
Thank you. I'm looking back through my code and there are really a lot of indicators. It would get pretty technical to explain them. They are all explainable it's just that each one corresponds to slightly different market conditions and I just didn't want to get into it.
Fair enough. Most people considering trying this probably have a few ideas for indicators.
But for anyone coming to HFT from a coding background instead of a trading background, an explanation of one of your indicators would have been fascinating.
Re: How I made $500k with machine learning and high frequency trading
#196Isn't profit meaningless without knowing initial investment? I read the first few paragraphs and got bored. Why not say upfront what the bankroll was to start?
edit: I found what I was looking for in the comments.
Re: How I made $500k with machine learning and high frequency trading
#197The charts show he was trading between Jun 2009 and Oct 2010. How much of his gains could be attributed to the market recovery in general? The Dow went from about 7000 to 11000, the Russell from about 600 to 800.
That's about 50% gain. Unless he put in $1m that doesn't completely explain his ROI.
Considering he was making 1000-4000 trades per day, I assume he put more than $1M. Trade fees alone would eat a huge chunk of money. The cheapest trade I can find is $3.95 (optionshouse.com).
So the fees will be $3950 - $15,800 per day.
$15,800 - $79,000 per week.
Re: How I made $500k with machine learning and high frequency trading
#198This is just glorified gambling. I am not sure what special insight or advantage he had, other than his own model. Every trader has a model. It could have easily been called "how i lost 500k with machine learning". Like gambling, it's easy to manipulate statistics to show that you did well in some period of time. I worked for a large investment bank about 10 years ago, writing trading programs for quant traders who w…
You know, that is a really good argument. Except two things: 1) He didn't lose money, he made 500k. 2) If this worked reliably, you would be out of a job.
Er... it doesn't work reliably, and by definition, cannot work reliably. If market inefficiencies exist to be exploited, then someone is ultimately getting the short end of every stick.
Re: How I made $500k with machine learning and high frequency trading
#199Out of interest, how much capital did you start off with?
$10k loan from dad. Built it up to 30k trading manually before my automated program went live.
What were your average transaction costs per trade? 1000 to 4000 trades per day, lets say 2500 on average, translates to about 625K transactions per year. I assume you did not have to pay something like ETrades 3$ commission per future contract, which would result in almost 2M of fees per year.
Re: How I made $500k with machine learning and high frequency trading
#200Earlier quoted context omitted.
Not really. A lot of people in the business would pay e.g. $5,000 for exclusive rights for something that worked this well in 2009 (with proof that it worked in 2009, e.g. verifiable broker statements), and a smaller amount (say, $5,00) for non exclusive rights. If he claimed it still works but he wants to sell it, it is a completely different game -- because when these things work, they are cash cows.
Yes i'm pretty sure it wouldn't work today. By buying the code I realistically mean hiring me to work for them based on what I achieved.
The guys you want to work for (2sigma, RenTec, Jane Street, Susquehanna, ...) are unlikely to call you up as a result of this blog post / hacker news exposure/discussion.
If you want to go back to trading, you'll probably have to actively try to get a job -- at the very least, let someone who's still in the business know that you are looking. In my experience in this field, word of mouth and friends-of-friends are infinitely more successful hiring strategies, for both sides.
(Re:releasing the source - I would like to have a look at the strategy, but I would recommend against releasing anything that is even close to being useful, unless you want to spend the next year screening "where can I get a good XTAPI broker" and "I've got XTrader_PRO set up, but I'm getting error 10013, what gives?" emails).
the Nuclear Phynance message board is probably a better place to look for business offers.