I was thinking it would be an extreme in a dull market and that results would be more unpredictable in a volatile market. I know Don's strategy, got that ticking, but always looking for more!
What I don't understand about program trading is exactly what happens mechanically from the point when a trigger is hit until the return to FAIR VALUE. Let me use your example where a "premium" of $3 or more triggers a "buy program". So, if spot is 1000 and the futures get to 1003 automatic programs BUY S&P 500 stocks and sell futures. Now lets say the premium remains at greater than $3 for a full 90 seconds before falling back to the fair value of $2. Do you know if these automated programs are programmed to buy fixed quantities of stock just once from trigger until the return to fair value OR if they keep buying more tick by tick while the premium is above the trigger point (or maybe above fair value)? Do you know if there are any other finer points to Program trading in general? (eg: buying may only commence if the trigger point is exceeded for a duration of at least two ticks)
Cash, Yes, I really should have thought of this when I made the post to Bung about this strategy. In a gapping market, there is too much to see. (And also, as I said, I usually have other positions to be concerned with). So yeah, it is definitely the stand outs in a dull market open that are the plays. Sorry I just did not make that more clear. Peace, RS8.5
Good questions, and truthfully I don't have any answers that I can give with perfect certainty. When I did this kind of trading, we did not do it the same way as I described. We were not big enough (and we were plenty big....maybe in another post I will talk about how we did do it). But as far as the answers I would think are "safe" for you, in the case you give, where the premium stays high enough, I would guess they could keep buying stock and selling futures. I don't see any risk in this, because at expiration, the fair value will be at zero. As to the "finer points" question, yes, duration is definitely a factor. They need to recognize if there is enough volume on the buy side (futures) to be sure they can sell. Enough stock to sell as well. I am guessing here....(I haven't actually participated in this since 1990) that representative baskets of stocks are used more now than the actual total basket .... all 500 stocks. So many of them, especially now, are very low priced, not that liquid, and don't move enough and are not weighted enough to make any significant difference. Today's technology is so far advanced from 1990 that I am sure everything is much cleaner, simpler, faster. More accurate, and more efficient in every way. Look at our own technology. And costs. As a market maker, I paid First Options (our clearing firm) almost $2 a contract to clear our trades on options. (thousands of contracts a day). Today anyone can get a better rate at home. Our Reuters quote machine that did simulations was an extremely hot machine (Shwartzatron)..... a very expensive 386 processor with a math coprocessor. And I think 2 or 4 megs of RAM. Clock speed maybe 33Mhz.? Can't remember exactly, but it was hot stuff then, and you wouldn't use it as a paperweight now. Peace, Rs8.5
The obvoius place to look eh.. 8 pages on fair value including formulas for calculating interest rate/ time decay effects http://www.cme.com/allaire/spectra/system/securemediastore/fairvalu.pdf
Very true: way over my head. I liked your explanation a lot more. It was understandable to this idiot too. Thanks again RS. I am surprised that anyone has voted your post as "worthless". Even if they understood all of everything and more, how could it be "worthless" if it helps what seems like 90% or so of the poll's voters to have a better understanding? Michelle
Yes, a lttle bit too much information. RS's explanation is a lot more understandable. Guess I am another one of the "idiots" here too. This link doesn't seem to have much to do with what I could possibly use in my trading. Guess if I had a few more million to trade with, it would be more interesting. MarcD
Thanks for your comments Rs8.5. Also thanks joethemoustache for the cme fair value link. I agree its a bit confusing. What I find interesting is that these trigger levels that places like www.programtrading.com publish are not really that hard and fast, as you will get more participants the greater the premium/discount. I suppose that is obvious in a way, but it does highlight the fact that different players face different cost structures (interest and commissions) and so face different trigger points at which arbitrage becomes profitable. The article also suggests that program buying should be more common than program selling because of the uptick rule (slippage or cost of conversions/bullets). I haven't followed this that long but the envelope of the published trigger points doesn't seem to support that.
What am I missing? The fair value has been between -2 and +0.5 for some weeks now. Why is it not a number (for the June contract) like 10 that slowly converges to zero at expiration? (June expiration for June futures contract)