Journey from investment bank to independent automated trader
- Thread starter lolatency
- Start date
What do you think is easier to predict, what is going to happen in the 50 milliseconds or what is going to happen in the next 1 year??
They are both equally impossible to predict. The best traders don't try to predict anything - they follow their system.
Tell you what though - I'd much rather be following a 1 year trend than a 50ms trend for the same 1 unit of trading costs.
They are both equally impossible to predict. The best traders don't try to predict anything - they follow their system.
Tell you what though - I'd much rather be following a 1 year trend than a 50ms trend for the same 1 unit of trading costs.
This is actually false.
Granted, I am NOT saying anything can be forecast with 100% accuracy, but rather with positive expectancy in statistically significant way, yes.
Fyi the market completely disagrees with your view right now. Banks and hedge funds are throwing money into high frequency and ultra high frequency systems, but according to you they would do better following the 1year trend. So in your world they obviously are wrong and they for some reason or the other are throwing tens of millions at things that are a complete waste of time, right?
Do you really think that you are so far ahead of some of the top hedge funds and IB's in the world???
What is your _realized_ sharpe/sortino for '08/'09 out of curiousity?
Ever tried to raise institutional money with your systems?
Transatlantic
I was hoping not to be put on the spot like that, but if you insist, yes, I'm some way ahead.
The problem at the heart of any investment bank is the compensation structure and Naseem Nicholas Taleb's by now (in)famous Black Swan principle - these guys want nice bonuses year in year out, and management tells them to make nice "steady" profits because that makes them feel good. Hence we have all this low latency, millisecond trading stuff in the desparate struggle to improve "predictability" and get those longed for steady returns.
When who ever it was said he wanted to make $50 per day to pay for groceries I had to resist telling him "why not aim to LOSE $50 per day and buy all your groceries on the few odd days you make $5000?"
Long term trend following is where the money is. Most - even the funds I regard as not particularly good for a variety of reasons - did fantastically last year as markets crumbled to dust over the course of a few weeks. The big money comes from the big moves - which no one can ever predict, merely prepare themselves for.
I don't personally manage any money - I'm just a client of a fund, albeit a very smart client who knows how to alter trading size contingent on system performance. I started 2008 with $100K and made around 100% by Dec 31 with a 15% worst drawdown.
I am 25% down YTD for 2009. Am I worried? Hell no, but if I worked for an iBank I'd be out on my rear. And by the way it gives me a lovely warm glow inside to know that the big institutional money disagrees with me
May I suggest perhaps you are not as far ahead of the some of the top hedge funds and IBs in the world as you might think. LOL
Don't make the mistake of thinking all institutional money = dumb money. That would be very sloppy thinking.
Transatlantic
I agree that you want to go as high frequency as possible, given you have an edge in this game, and so far I still believe (unless otherwise proven) that the biggest edge is from seeing order flow. If you do not possess such edge I think that an edge can be obtained from a solid longer-term trend following strategy and in fact thats what I myself specialize in. Many critics often assert that 80% of the times markets are non-trending which is correct. But I believe that is a very flawed way of thinking. You do not get paid in time but you are paid in dollars, and fact is that the market is taken from one price level to the next in what could be termed a trend. So, the biggest moves are on the back of trends. I do not define trends in terms of duration of trade but in terms of amount of cointegrated price moves. Of course simple cross-overs are totally naive and it can be empirically shown that there is no edge.
Just my point on your earlier remark but I wont continue as the thread is not about trend following.
Just my point on your earlier remark but I wont continue as the thread is not about trend following.
Quote from Transatlantic99:
This is actually false.
Granted, I am NOT saying anything can be forecast with 100% accuracy, but rather with positive expectancy in statistically significant way, yes.
Fyi the market completely disagrees with your view right now. Banks and hedge funds are throwing money into high frequency and ultra high frequency systems, but according to you they would do better following the 1year trend. So in your world they obviously are wrong and they for some reason or the other are throwing tens of millions at things that are a complete waste of time, right?
Do you really think that you are so far ahead of some of the top hedge funds and IB's in the world???
What is your _realized_ sharpe/sortino for '08/'09 out of curiousity?
Ever tried to raise institutional money with your systems?
Transatlantic
You prove my point exactly. You think I'm not ahead this year (ok let's forget 2008) because you think making money every quarter or month matters. The idea that -25% is a price you pay, or upfront investment, to make really big money later on hasn't occured to you. I really don't think you understood what I said.
There are in fact very few large funds falling outside the TF category with any significant smarts. The true long term performers, such as Tudor and Renaissance, have been completely closed to new funds for decades, and hence they are not institutions in my book.
I would be grateful if you could suggest some smart institutions. I will then do the world a favour and recommend them on CNN money
There are in fact very few large funds falling outside the TF category with any significant smarts. The true long term performers, such as Tudor and Renaissance, have been completely closed to new funds for decades, and hence they are not institutions in my book.
I would be grateful if you could suggest some smart institutions. I will then do the world a favour and recommend them on CNN money
Quote from Ash1972:
I was hoping not to be put on the spot like that, but if you insist, yes, I'm some way ahead.
The problem at the heart of any investment bank is the compensation structure and Naseem Nicholas Taleb's by now (in)famous Black Swan principle - these guys want nice bonuses year in year out, and management tells them to make nice "steady" profits because that makes them feel good. Hence we have all this low latency, millisecond trading stuff in the desparate struggle to improve "predictability" and get those longed for steady returns.
When who ever it was said he wanted to make $50 per day to pay for groceries I had to resist telling him "why not aim to LOSE $50 per day and buy all your groceries on the few odd days you make $5000?"
Long term trend following is where the money is. Most - even the funds I regard as not particularly good for a variety of reasons - did fantastically last year as markets crumbled to dust over the course of a few weeks. The big money comes from the big moves - which no one can ever predict, merely prepare themselves for.
I don't personally manage any money - I'm just a client of a fund, albeit a very smart client who knows how to alter trading size contingent on system performance. I started 2008 with $100K and made around 100% by Dec 31 with a 15% worst drawdown.
I am 25% down YTD for 2009. Am I worried? Hell no, but if I worked for an iBank I'd be out on my rear. And by the way it gives me a lovely warm glow inside to know that the big institutional money disagrees with me
Perhaps you, yourself, have been fooled by randomness?
Me fooled by randomness? Well, it's not impossible.. any system that is set up to profit from big moves could be susceptible to no such moves happening for a long time - possibly ever. The lack of black swans for longer than you expect could itself be a black swan.
However, there is a mountain of evidence to suggest that those who want to make money frequently by betting on likely events (i.e. betting against unlikely ones) will ultimately lose. Trying to "get it right" all the time is precisely the problem. Those who bet ON the unlikely outcomes are the big winners.
So few people get this it's staggering. Can we have a show of hands? Who actually agrees with me?
However, there is a mountain of evidence to suggest that those who want to make money frequently by betting on likely events (i.e. betting against unlikely ones) will ultimately lose. Trying to "get it right" all the time is precisely the problem. Those who bet ON the unlikely outcomes are the big winners.
So few people get this it's staggering. Can we have a show of hands? Who actually agrees with me?