Program Trading, Arbitrage, Ect...

[darthtrader]

Anyone know much about this stuff? Considering the % volume of program trades I'm trying to get a more general view of the mechanics.
What peaked my interest was this little pdf that was from a presentation at the futures summit 2 years ago:
"Anti-Arbitrage" - Combating Automated Trading Systems
http://www.fts.cme.com/files/janowski.pdf
I would love to have checked out that presentation but I can't find anyway to access it now.
I know I've read somewhere that the arbs use the ES to "sweep" the big pit contract. I can only vaguely understand what that means. S&P is a good example. You have cash, the big pit contract, SPY, ES...anyone have an understanding of how things fit togather?

[Soultrader]

I can explain a few things from what I know here in Japan. I dont know much about arbitrage in the ES and such but for the Nikkei it is rather simple.

The Nikkei is traded both in the Osaka Stock Exchange (OSE) and Singapore Exchange (SIMEX). The tick increment for the OSE contract is 10 yen while the SIMEX contract is 5 yen.

1 tick (10 yen) = approx $100 for OSE contract
1 tick (5 yen) = approx $25 for the SIMEX contract

So arbitrage occurs simply by buying the bid and selling the ask. For example, you will buy 2 contracts at SIMEX at the bid of 12755 and then offer 1 contract at the ask on the OSE at 12760. This arbitrage allows a 5 yen profit with minimal risk. Unfortunately the system and resources required to do this can only be done by institutions.

In regards to program trading, I have no clue how the calculations are done but depending on the difference btween the cash and futures, institutional traders will send out basket orders. Early morning reversals here in Japan are usually program orders. This is a common strategy applied by many institutions here. We used to be amazed by the 10:00am reversals as it was most likely controlled by one big institutional trader. What this guy used to do was buy a basket of best performing stocks and sell a basket of worst performing stocks. This caused massive reversals in the futures market.

[darthtrader]

ahh good stuff. Little stories like that are what I'm looking for, just little insights.
I guess in general what I would like to eventually get into is tracking a few highly correlated instruments against YM as an indicator. Not to do any sort of arbitrage but as that pdf above mentions, "the only way to beat the programs is to not trade with them".
I kind of wonder how many times you have a valid setup on your instrument that would have been a winning trade, but some other correlated instrument has moved out of wack(that your not even thinking about) so the programs/arbs come in and mess things up with an arbitrage. Would be interesting then too to see if your setups worked better when correlation was tight with nothing for the programs to trade.

[thrunner]

I don't calculate this stuff, but this may be what you are looking for at Program Trading, Fair Value, Index Arbitrage Values - indexArb.com webpage : sample Esignal calculation efs.

	Quote:
	Q: What is program buying and selling? A: When the S&P 500 premium (the difference between the S&P500 futures price and the S&P500 cash (regular index) price get to certain extreme levels (as identified by the buy and sell thresholds on indexarb.com for example), a certain kind of program trading (computerized trading) will take place to take advantage of the small out-of-whack premium levels: index arbitrage. Program Buying - When top circled premium is hit institutions buy the S&P500 stocks and short the S&P 500 futures. Program Selling - When bottom circled premium is hit institutions will sell the S&P500 stocks and buy the S&P 500 futures. In short program buying and selling do have an effect on the market, but it is not always clear which way the market will be influenced. If you hit the program sell threshold, the S&P500 futures (and thus the DOW) do not necessarily move up, and vice versa with the buy threshold being hit. There are entire courses and systems/services dedicated to determining which way the market will move based on program trading.

[mister ed]

Hi Darth and Soul

I am not sure if this is the sort of thing interesting to you also Darth - I have cut and pasted a few articles from around the place on algorithmic trading. Might be of interest (I find this stuff fascinating ... and try to learn about it on the basis that the more I know about it the better I can understand whats happening in the market).

All the articles are referenced with links so if you want to check out the original etc.

[darthtrader]

Ahh great stuff guys, I remember that indexarb site but couldn't find it tonight.
I found this site that has some interesting information from that pdf somehow, haven't gone through it all yet:
SPI Futures trading systems components of SPI200 index futures

I totally agree Mr Ed, its extremely fascinating although I'm still trying to figure out if there is any utility to this stuff. I read some on some site from that pdf that some of the program execution trades have got down to 10 milliseconds and what an improvement that is over 100 milliseconds....

[darthtrader]

so does anyone use PREM still here? I remember soultrader had it on his videos but I didn't really get it then.
I've been watching the fair value levels from indexarb all morning on YM and S&P PREM. I almost want to say those levels are kind of bogus. 3 hours in and the PREM has not hit the level they gave any probability of program trading to yet on either direction. If programs are such a high volume of the total market that makes no sense at all. I almost wonder if their formula hasn't accounted for the dramatic increase in program trading in recent years.

Something interesting from watching the tape this morning is when PREM started moving up, once it hit its peak above 2 you seen the tape go crazy and a fast 20 point move up, closing and prem go back to being flat. I would think that was basically programs arbing the ES/cash and then arbing on the YM/cash dow to keep everything in line. If that was the case then PREM should be a pretty nice thing to watch. Especially if your looking to counter trend trade, doesn't make much sense to be on the wrong side of PREM if the programs are going to come in and blast you with heat on the otherside to close the gap with cash.

[scalper]

	Quote:	Soultrader
	In regards to program trading, I have no clue how the calculations are done but depending on the difference btween the cash and futures, institutional traders will send out basket orders. Early morning reversals here in Japan are usually program orders. This is a common strategy applied by many institutions here. We used to be amazed by the 10:00am reversals as it was most likely controlled by one big institutional trader. What this guy used to do was buy a basket of best performing stocks and sell a basket of worst performing stocks. This caused massive reversals in the futures market.

interesting information.. Thanks

[mister ed]

	Quote:	darthtrader
	If programs are such a high volume of the total market that makes no sense at all.

I think 'program' trading, in a narrow sense, refers to those arbitrage activities where stock bought/futures sold or stocks sold/futures bought, whereas there is also a broader sense in which the word is used to refer to other types of automated activity. I think the best word to encompass all the types of automated activity executed by computers/robots is algorithmic trading. Wikipedia, while not a definitive source, has a good selection of articles and links about algorithmic trading.

Back to program trading and the phenomenon you described Darth. My understanding (poor though it is) of program trading (in the narrow sense described in previous paragraph) is that it is purely arbitrage, and it will account for some (quite a few) of the short, sharp moves in the futures, bringing the cash and futures 'into line'. I think an extended move like the one you described is not arbitrage but more of a directional strategy? I am out of my depth here but I think we can uncover some valuable information if we continue on with this thread. I like your attitude of trying to find utility in this stuff too, let's not just understand this stuff, lets use it to our advantage.

Also, I am looking forward to the movie 'Terminator 4: Defeat of the Algos'.

[HAL9000]

	Quote:	mister ed
	Back to program trading ...

Excuse me if I invite myself, but in my opinion, you have to choices (as nearly always) if a program signal appears: Join it, or fade it.

Well, the PREM stuff makes sense to me, even if I don't use it.
BTW, for a long time I worked on algos, but currently I like using my brain, one of my fingers, for a click.
But at some point I will reach my goal with algos.
Well and I see algos from others, just don't ignore them.

Hal

P.S.: IMO we are already watching Terminator 5: "If you can't beat them, join them!"