How do big players trade?

Read:

Inside the House of Money.
The Invisible Hands.

Both by Steven Drogny. He interviews fund managers, asks about their favourite trades etc. [If you're creative they're both available free on the net].
 
They don't trade, they position.

Great comment.

The biggest headache for these guys is managing these positions without moving the market against them and incurring a liquidity premium.

The attached article is a beautiful exposition of the tradeoff between immediacy and lower transaction costs.
 

Attachments

  • Optimal control of execution costs.pdf
    671.4 KB · Views: 771
That is freaking brilliant, Sang. I wonder how many people are enjoying that kind of stress level from the home!
 
who really knows? I bet they have a magic parrot. But what is the point of knowing? Do u have hundreds of millions of dollars to invest?

Also where does one find a magic parrot:LOL::LOL::LOL::LOL::LOL:
 
I get the feeling they usually trade like this:


Leeson is interesting, he was smart in terms of being able to pick up on weaknesses within Barings; a small bank with relatively few checks, an arrogant attitude (they had been bailed out about 100 years before Leeson brought them down) and a public schoolboy culture to boot. So the fact he was in control of the front & back offices meant he could do what he did. However he was dumb in terms of the bank he chose, he had left JP Morgan, what ultimately brought him down was the fact that Barings didn't have the capital to finance his margin payments; if he had been at a bigger bank with more capital he could probably have supported the price and slowly sold off his position after the effects of the Kobe earthquake had died down.

What I didn't understand about his crime and the collapse of Barings was the fact that Barings didn't seem to be quite as connected as it thought it was, can't banks borrow massive sums of money off one another at very reasonable rates? Why didn't Barings just borrow however much Leeson needed to be able to meet his margin payments without having to sell part of his position thereby bringing the price down?
 
"The regulatory authorities were poor. On a daily basis, I could represent between 60 per cent and 80 per cent of the daily volume on SIMEX and at least 40 per cent of the open interest. These levels of trading should have attracted more of the exchange's attention. If anything, they attracted less. There was a SIMEX awards ceremony in 1994 when the management came out to receive the award for the top customer for volume on SIMEX. The 88888 accounts probably represented 95 per cent of that customer volume. SIMEX [also] paid 100 per cent of the option premium during this period. That's not usual. It doesn't happen in most markets. If they hadn't done that, I wouldn't have been able to do what I was doing.

The Bank of England also messed up. The capital of Barings was only £250 million and I had £500 million of it. I know I failed my math A-level, but even I could add that sum. They allowed Barings to send over 100 per cent of the bank's capital to me in Singapore when the legal limit was only 20 per cent. [Barings] Treasury used to make me daily payments of between $50 million and $100 million without any logical reason".



...........................
Just staggering hey.

Read the whole interview here: http://flirtingwithdisaster.net/the-collapse-of-barings_284.html
 
I am talking about repeated patterns in the way that price moves and behaves.

This can be seen in a chart but it has nothing to do with common chart patterns.

Exactly. This is how Prop Trading works at IB's and Hedge Funds. One friend of mine is a Quant Trader at Merrill Lynch and informed me that most of the traditional Prop Traders have been sucked and they are only implementing "Stat Arb" strategies. The same applies for Goldman Sachs. That is partially because of the Paul Volcker rule. For those of you interested in this topic please look for the terms "Machine Learning", "Signal Processing" and especially "Data Mining" or "Data Snooping".

With regards to Data Mining I am in better position to comment on those strategies. That is because my Masters Dissertation was in Data Mining. In particular, I investigated and provided evidences of various Seasonality Patterns in the Greek Stock Market over a period of 20 years (The sample was 4,983 observations!!!, so it was a very difficult task to accomplish). Those seasonality patterns included the "Day of the week Effect", "January Effect" and "Holiday Effect" (google them),which were persistently present in the Greek Stock Market.

For those positions, the qualification required is a PhD (or MSc in the best case scenario for you...) from a Top Tier uni (Oxmbridge, UCL, Imperial, King's College...) in a highly quantitative subject such as Maths, Engineering, Physics etc... So, it is not easy at all to get that kind of positions....

However, caution is required. Those strategies , most of the times, will not work under all market conditions. It is usual to observe that some of those models that have performed reasonably well in the past , will not have the same performance at present and/or in the future. This is well known to Quant Traders/Analysts. Please refer to this acedemic paper:

http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.12.1040
This is by far one of the most exciting and useful papers I have ever read!!!!(y)(y)(y)

The most difficult task is to identify the period that this model will stop working and as soon as you identify this period to stop using this model...This friend of mine also told me that one of the most successful models they were using before the Credit Crunch was resulting in heavy losses under the altered market conditions of the post 2008 era... They stopped it and now that the market conditions have more similarities with the period prior to the Credit Crunch , they started using it again and it currently generates profits...

I hope my post has been helpful.:clap::clap:
 
Read the recent New Yorker article about Ray Dalio and Bridgewater... There's some interesting stuff in there.
 
Hi guys. Do any of you know how large institutional traders(the ones with the big money) think and trade? QUOTE]

With other people's money

:smart:

Exactly! The same principle applies to their salaries and bonuses (I am talking about IBs' traders and not hedge funds' traders). When they outperform they get BIG bonuses. However, when they fail they turn to the taxpayers to bail them out!!! Smart, really smart!!!:smart::smart:

I am not implying that they shouldn't get their bonuses when they are successful , but when they mess up, they should take the responsibility of their actions and face the consequences... And if they are not ready for the consequences they shouldn't act in such a manner. As every other Prop Trader... However, this has not been the case for most of them..(except from Lehman Brothers)
 
Where are you from, Oppi?

Exactly. This is how Prop Trading works at IB's and Hedge Funds. One friend of mine is a Quant Trader at Merrill Lynch and informed me that most of the traditional Prop Traders have been sucked and they are only implementing "Stat Arb" strategies. The same applies for Goldman Sachs. That is partially because of the Paul Volcker rule. For those of you interested in this topic please look for the terms "Machine Learning", "Signal Processing" and especially "Data Mining" or "Data Snooping".

With regards to Data Mining I am in better position to comment on those strategies. That is because my Masters Dissertation was in Data Mining. In particular, I investigated and provided evidences of various Seasonality Patterns in the Greek Stock Market over a period of 20 years (The sample was 4,983 observations!!!, so it was a very difficult task to accomplish). Those seasonality patterns included the "Day of the week Effect", "January Effect" and "Holiday Effect" (google them),which were persistently present in the Greek Stock Market.

For those positions, the qualification required is a PhD (or MSc in the best case scenario for you...) from a Top Tier uni (Oxmbridge, UCL, Imperial, King's College...) in a highly quantitative subject such as Maths, Engineering, Physics etc... So, it is not easy at all to get that kind of positions....

However, caution is required. Those strategies , most of the times, will not work under all market conditions. It is usual to observe that some of those models that have performed reasonably well in the past , will not have the same performance at present and/or in the future. This is well known to Quant Traders/Analysts. Please refer to this acedemic paper:

http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.12.1040
This is by far one of the most exciting and useful papers I have ever read!!!!(y)(y)(y)

The most difficult task is to identify the period that this model will stop working and as soon as you identify this period to stop using this model...This friend of mine also told me that one of the most successful models they were using before the Credit Crunch was resulting in heavy losses under the altered market conditions of the post 2008 era... They stopped it and now that the market conditions have more similarities with the period prior to the Credit Crunch , they started using it again and it currently generates profits...

I hope my post has been helpful.:clap::clap:
 
I am from Greece, but I live in London. I got my Bachelors degree from a Greek uni and I got my Masters degree from a UK university.

Ah, hence the Greek stock market. I thought you made some interesting observations in your post (y).
 
Ah, hence the Greek stock market. I thought you made some interesting observations in your post (y).


Well, I have been investing in the Greek Stock Market for a number of years but I have never had that kind of knowledge before. So, as soon as I got the quant skills required I used those statistical tools to get a better understanding of the market.

As mentioned, I provided evidences of various (quite a lot of them!!!) seasonality patterns present in the Greek Stock Market. This was a very strong violation of the efficient market hypothesis (I know most of you think that this assumption is BS). I was under the impression that the Greek Stock market was indeed not efficient at all but till that time I could not prove it. Also a very interesting observation was the period prior to 2002 and after 2002 (the introduction of the Euro). Some of those patterns vanished (after the introduction of Euro) and some of them changed to the same patterns as the other developed markets. So Euro did have some effect on the Greek Stock Market. However, some of the patterns remained the same (a very striking observation if you consider the time period- 20 years!!!).

I am currently working on my Dissertation alongside with my supervisor and we will submit it to be peer reviewed. We intend to publish it as a financial research paper. Let us hope so.....:cheesy::cheesy:
 
It makes sense, after all patterns are just a picture of price behaviour.

Well, I have been investing in the Greek Stock Market for a number of years but I have never had that kind of knowledge before. So, as soon as I got the quant skills required I used those statistical tools to get a better understanding of the market.

As mentioned, I provided evidences of various (quite a lot of them!!!) seasonality patterns present in the Greek Stock Market. This was a very strong violation of the efficient market hypothesis (I know most of you think that this assumption is BS). I was under the impression that the Greek Stock market was indeed not efficient at all but till that time I could not prove it. Also a very interesting observation was the period prior to 2002 and after 2002 (the introduction of the Euro). Some of those patterns vanished (after the introduction of Euro) and some of them changed to the same patterns as the other developed markets. So Euro did have some effect on the Greek Stock Market. However, some of the patterns remained the same (a very striking observation if you consider the time period- 20 years!!!).

I am currently working on my Dissertation alongside with my supervisor and we will submit it to be peer reviewed. We intend to publish it as a financial research paper. Let us hope so.....:cheesy::cheesy:
 
"The regulatory authorities were poor. On a daily basis, I could represent between 60 per cent and 80 per cent of the daily volume on SIMEX and at least 40 per cent of the open interest. These levels of trading should have attracted more of the exchange's attention. If anything, they attracted less. There was a SIMEX awards ceremony in 1994 when the management came out to receive the award for the top customer for volume on SIMEX. The 88888 accounts probably represented 95 per cent of that customer volume. SIMEX [also] paid 100 per cent of the option premium during this period. That's not usual. It doesn't happen in most markets. If they hadn't done that, I wouldn't have been able to do what I was doing.

The Bank of England also messed up. The capital of Barings was only £250 million and I had £500 million of it. I know I failed my math A-level, but even I could add that sum. They allowed Barings to send over 100 per cent of the bank's capital to me in Singapore when the legal limit was only 20 per cent. [Barings] Treasury used to make me daily payments of between $50 million and $100 million without any logical reason".



...........................
Just staggering hey.

Read the whole interview here: http://flirtingwithdisaster.net/the-collapse-of-barings_284.html

Yeah, the poor regulation seems to be a recurrent feature in these cases, but Leeson was undone by Kobe & a lack of capital. Even though Barings were pretty easy to fool they just didn't have enough money to support the market in the event of a serious natural disaster.

I just wonder how much Leeson would have made had he been able to keep the prices as high as they were at their peak, the figures he was submitting at one point suggested he was making capital gains of £10 million a week (how they believed, if they really did, that an arbitrageur was making those sort of returns is beyond comprehension).
 
Seasonal effects in stock mkts are a much discussed subject. Lots of research on it. I don't know if it violates EMH, per se, assuming you're talking about weak-form EMH.
 
Top