A few comments on the why's? At my undergraduate alma-mater, above all else, we were required to think for ourselves. Question everything and seek our own truths. Class (well the humanities one's at least) were conducted to gather everyone's thoughts and perspectives. In the markets we have to deal with the truth of the matter. Somewhere recently, Grob posted a diagram of all the market participants; two of them being arbitrageurs. Although my market experience is substantially shorter than Grob's, I will comment on what it is that is operating when things SQU/NEUT/STR. Arbitrageurs play a large role in neutralizing the offset which is why I mentioned using DDE as opposed to the 6 second snapshot that alot of other datafeeds provide. Arbitrageurs are a large group who's influence diminish as the offset neutralizes. Everyday, the arbitrageurs have their fair value numbers plugged in and ready to go much like we do. Their trigger levels vary over small range. When the SMART MONEY folks STRetch out the futures by some critical value, arbitrage programs automatically kick in to pocket the offset differential (ie. SHORT the futures, LONG all the underlying stocks of the index) and vice versa. As you can imagine, synch is their biggest single trade money maker for the day. I have seen synch offsets STRetched out 30+ before reaching a synch of ZERO. When the offset NEUTRALIZES, the arbitrage programs close their position. Were I them, I would save on costs by SCTing this offset (ie. {STR=REVERSE/SHORT THE OFFSET}<=>{NEUT=HOLD}<=>{SQU=REVERSE/LONG THE OFFSET})... LOL... SCT is everywhere... Acquiring the whole index at a single instant is difficult at best. Most arbs will simply concentrate on a subset of the indices' stock to do their arbing. It is a bit more risk but as one would expect, also more rewarding... Take any futures that is tied to an index, and the above is the truth of the matter. Kind Regards, MAK
Mak, one of the points Jack makes on the market log drill is to determine the extreme values of str/squ (i.e. non neutral) that cause price change. I got the impression that he viewed the neutral range as dynamic from day to day needing recalibration of the boundries, but your spreadsheet has neutral fixed at +/-2... do you find yourself making mental adjustments to the readings? Example maybe you see +2.5 reading str but you know today neutral is closer to +/-3?
I have been playing with NQ since they made the announcement. The spread advantage of YM with the volume of NQ. I think it was a brilliant move. Ireland You rock bro..... p.s. The trolls are a small price to pay for what is available.
Great Q... On occasion I have been tempted to adjust... One of the primary objectives of the spreadsheet was to specifically isolate SMART MONEY movement of YM and it's highly correlated ES counterpart. Some of you will catch my drift on where that can go. The +/- 2 is simply the results of many many days of personal debriefing. Many versions back in the sheet's development, I used to bound (explicit match) the ranges of the pair to each other (ie. YM.HI<=>INDU.HI and vice versa for the LO). The reasons for it were basic. There were a ton of noisy components which had to be filtered out. For example, using tick increments for the range helped tremendously as opposed to dynamic increments. After many nights of rehashing what was what, I realized that all that was needed was window by which one could monitor and be convinced that "YES, I see that it is true that the YM/INDU pair lead the S&P". I have run many iterative refinements on the sheet and finally settled back on the +/-2 as being not too loose nor too tight, and very effective with 2 pairs. As you improve your effectiveness in using STR.SQU, you do run into the occasional shifts from +3 STR and -1 SQU as opposed to the +2 STR and -2 SQU. For me, when I find that this occurence is sustaining itself, I simply increase the offset cell by 1 so as to RE-CENTER the NEUTral zone. In other words, when +3 STR and -1 SQU are non-noise signals, then I would take an original offset value of say 59 and make it 60 so as to recenter the STR/SQU to +/-2. The other way to do it is to actual change the blue celled offset values found on the left side of the sheet. Had I spent a little more time, I would have adjusted the OFFSET range to adjust precisely off the two blue colored cells. This detail which you bring up, is a rather fine fine-tuning of the action. As with my channels, I just keep loose and stick with my +/-2. Adjust it to your preference. The majority days, I haven't found a reason to recenter. We are all calibrated differently visually and that is OK. Hence the flexibility of the sheet. As far as the mechanics behind the highlighting etc, they are simple Conditional Formating formulas which can be viewed by clicking on a cell like STRETCH and then going to the toolbar at the top and selecting FORMAT->CONDITIONAL FORMATING. There is nothing terribly excited about anything on the sheet and all the content there is 100% transparent. It is simply a bare bones tool that gets the job done.... Perhaps some of you are still working to get it situated. If it doesn't fit in to your current level, just save it for later. We are all at different places. For example, I am still trying to recognize (ie. get calibrated to) and bridge GAUSSIANS to everything else. Ireland, thanks for the IB equivalent. Kind Regards ALL, MAK
This is pretty cool. A lot of noise but occasionally a heads up like just now. ES was basing but STR was sticking and a few moments later pulled ES down. Sticking is a nice clue when it happens. I would think that someone could scalp pretty effectively it they took the time to learn this stuff. Do you watch T&S mak? It seems to complement things pretty well.
A brief summary comment. Always take the time to do the grunt work to understand the measure at hand. then that in hand you can "see". The consequential automation then places at your figure tips the power to perform effectively and efficiently. MAK is the wave of the future for the "community" as we all see and admire. so now we are able to see and delve into the nuances of market moves since we can see tham from just before they begin by using a vernier fine control. We were fortunate that Chicken gave us all those refining efforts that were in play from a crude baseline (30 minute stuff). Here we are now looking at how to recognize the metrics of moves. The Nike I and Ajax systems solved two quadratics (target and missle) using an analogue array of a bunch of DC amplifiers. Two things made it possible: a dither oscillator (mechanical on the bearings of the antennas to take out starting friction and "zero setting the capacitors in the DC amplifiers over 200 times a second. The missle was detonated within the target, a millisecond problem. So now we see the market by doing transient analysis. We are using the Cray results of quants as our stepping stone. all we need to do is have our human reaction time operating as an anticipator of the sequences. The .1 and .5 stuff is child's play in some senses. But the metrics aren't Metrics give you the definition of the ball part and then you march from there. The answers to BO/FBO and pace are what have arrived on the table. So it is going to be fun. "Discoveries" come in the form of tools that are of utility for "seeing". As MAK extends vision, effectiveness and efficiency are iteratively refined. We all know by now that the equity curves blow up (not out) beyond normal distributions. Use time wisely to get a good advantage is base capital. The payoff goes way past being rich. I agree losing riskarb, and scientist et al is a very poor strategy for the management of ET to have designed. It makes you wonder why there isn't a money management forum as yet???? lol. Market theory? The chart (figure 8-1 out of Larry Harris is a good one)
For the last fifteen minutes I have been running the IQFeed and IB spreadsheets side-by-side. The vast majority of the time they are step-in-step, within a split second. But I did notice occasional lag on the IB, which could cause for missing some of the extremes. Example, both show a squ of -3, then IQ quickly jumps to -3.4 and returns to -3, while IB sticks on -3.
This is a amazing. ES has been forming this ascending triangle, trying its darndest to go up but the whole time we have been stuck in extreme squeeze. I began to think my program was stuck but es finally caved in. I see some real possiblities here. edit: During this whole down move sqeeze has ruled going to neutral occasionally, only a blip or two of green. Will be curious to see how this ends.