Search results

2004

Average is about 18%

A solid 20% consistently ? ;)

So lets say youre making a solid 20% consistently and that you have a starting capital of 1m usd. Now lets say you manage 250m$ with the same return and get 20% of performance fee Basic math would tell you that youre wrong

I would suggest to you Futures. I came into finance with a strong computer background and little financial background. - Forex is an OTC product so you will need to deal with a lot of issues. - Stocks is not centralized around a single exchange but rather through a variety of venues and...

Yes this is basic quant work, I have interns who do this better than coffee. I'm not saying that random numbers will go to 0, but the strategy of randomly buying and selling on a **real** market data set will not output any positive performance that passes statistical significance test. This...

Disagree with you too here. The us stock market is sold to the retail american as a safe capitalistic way to increase the productivity of the country. If it ends up being just a crooked casino youre losing a big market participant : retail investors.

Dandomnly generating data "looks" like the market but its not. Easy to prove. Simulate a 50/50 random buy sell every 1 minute on any market you like and aee if you end up at 0. Second option calculate entropy of your dataset

Thats not how you use machine learning. Thats just basic historical simulation.

I disagree with you. I've seen a lot of how equity hedge funds work in the US and the vast majority of them beating the index is based on insider information. Similar or not worse happens in fixed income funds. Until a few years ago you even had brokers of information you could pay 250k to talk...

Yes machine learning is what i use for the totality of my book

Oh ok cool. Learn c sharp, python is good for academics and basics but you still need a real programming language as its hard to solve complex issues fast with a single threaded language like python. Machine learning is just a component for system selections, focus on algo design first and...

Machine Learning can forecast anything from volatility to how long it takes me to make spaghetti. The question of "reliability" is more important when dealing with financial data.

You can start with Coursera's course on Machine Learning which will take you 12 weeks, assuming you have basics of calculus/algebra (otherwise start with calculus course that takes 8-10 weeks). If you have no knowledge of programming that's probably going to be the hardest, count a good 2-5...

Best is if you ask them directly. I used to pay 7k€ per month for collocation, which included a 2U server and maintenance in NY4. I've never done HFT on shares but on futures I used to burn 25k€ a month between cross-connect fees, failed-over servers etc. You may get away with 2k€ per U, but...

To be honest it's not that much about the money, but it's impossible to have access to the same infrastructure in a hedge fund and in a basement. Anything from data to simulation to crunching / machine learning infrastructure is just way too expensive even if you have a capital of 4-5m$. One of...

I agree it's not easy. Intraday the best I can do is 180$ on Crude Oil. I've stopped trading anything below 50-80$. Some markets are easier than others, for instance Wheat, Soybean, Cocoa, I've never ever ever been able to get > 40$

Youve noticed the fairly small average trade profit right ?

Oh ok sorry then if you backtested it should work out well. I know a 300m hedge fund that does only spx futures weekly options writting and they dont have too much liquidity issue but thats kind of the limit so you should be good.