Question for Grob/Hershey...

Status
Not open for further replies.
Quote from makosgu:

If you're a beginner, there is plenty to feast on in just the EXTREME arena alone. There also seems to be some other nags with just plain sitting on hands when wanting to be in and doing something. It is a nagging problem since the precursor for getting in is knowing what is going on. Knowing that you don't know what is going on and sidelining is acting directly in accordance with your game plan. Perhaps it is better for some folks to just TAPE OUT FAST/SLOW/MEDIUM/etc...

MAK
More...

I have participated in a number of these threads cocerning Jacks methods and it is always the same. Instead of starting at the beginning people tend to jump to the middle and endgame because it is more interesting. Jack has said many times that you should begin by tradng rockets until you have tripled your account. Does anyone do this? Not that I know of. Some people attempt rockets for a week or two then give up and move on. It took me over six months to understand rockets and how to trade them. Jack calls them no risk trades and they can be if you hang in there long enough to figure out where and when. Rockets are my bread and butter. I am trying to learn the slow paces but find it pretty difficult but I am in no hurry and it doesnt really matter if I ever get it.
 
I'm not sure I would be so quick to disregard ES/SPX spread. You may need to get the vertical scale right to see the movement - but it's there.

In watching it in conjunction with YM/INDU I notice that they often move pull in different directions - e.g. YM/INDU suggesting strength while ES/SPX suggesting weakness. This is typical during CCC. The spreads are more likely to move together when price movement is directional.

I can't say that I have a reliable way to use this information, and it may be redundant - just adding an observation.

1c
 
I have been watching the YM/INDU side by side for four months now. No hard data to back it up but I have noticed when the cash leads, the move seems to be more powerful and sustainable. Read on a thread this weekend where another trader has noticed the same exact thing. That was my reasoning a while back for monitoring the YM/INDU Str/Sq even though it's the instrument I trade. Trying to catch that move earlier with the help of a better visual. Due to screen limitations for the time being I just monitor the PRV tool while watching the YM/INDU screens....The new computer is much better but only as fast as the dial-up connection. Never shoulda moved out here in the swamp. :)
 
Rajin

Before we got dsl out here where I am I used a satalitte feed. The upload was still slow but the download speed was pretty fast and worked pretty well for data feed. You might want to check it out.
 
Quote from makosgu:

The dividend/divisor stuff does not vary during the day. The divisor only changes if components of the dow are swapped out (ie. SUN for MSFT). The dividend changes on the days dividends are paid out or reported for any of the components. When doing an RT fairvalue, what is important is to determine the two closest market points on the ZERO COUPON YIELD CURVE. Rollover is 31 days away so the points between the 1M LIBOR and the 3M LIBOR are the points to interpolate between as an example. I'm still refining to fine the actual yield curve points that indexarb uses since somehow, where they get there data is relatively spot for EOD. I presume that using the equivalent INTRADAY would be even better. It is work in progress since I still wind up with some differences...
More...

I've setup a realtime calc of FV to see how it compares through the day to the daily indexarb value. For the interest rate I'm taking the front two eurodollar rates and averaging the two, which seems to get me very close to the indexarb FV at the start of the day. But the one element of the equation that I do not have RT is the dividend component. If you look at indexarb's decomposition of the FV it seems that on many days the biggest changes to FV come from adjustments to the dividend component rather than from changes to rates or the index. Example, yesterday the dividend component (divisor adjusted) was 43.9, today it was 41.9. That is a two point adjustment to the FV caused by dividend announcements in a single day, and from what I've seen adjustments on this scale are common. When I started watching this daily it was a real eye-opener to see how much of an effect dividends have on the FV. Certainly the professional arbs are on top of these announcements and price them in ASAP.
 
Thanks Easy. I am going to check it out. The reason for the new computer was all the drop outs, etc....It's no better in that regard. Stayed sidelined all morning because my trading platform and chart data didn't jive. Very frustrating and happens all to often. Thanks again....
 
Quote from txuk:

I've setup a realtime calc of FV to see how it compares through the day to the daily indexarb value. For the interest rate I'm taking the front two eurodollar rates and averaging the two, which seems to get me very close to the indexarb FV at the start of the day. But the one element of the equation that I do not have RT is the dividend component. If you look at indexarb's decomposition of the FV it seems that on many days the biggest changes to FV come from adjustments to the dividend component rather than from changes to rates or the index. Example, yesterday the dividend component (divisor adjusted) was 43.9, today it was 41.9. That is a two point adjustment to the FV caused by dividend announcements in a single day, and from what I've seen adjustments on this scale are common. When I started watching this daily it was a real eye-opener to see how much of an effect dividends have on the FV. Certainly the professional arbs are on top of these announcements and price them in ASAP.
More...

Hmmm...

Some thoughts since you too are working hard at this. Definitely good to exchange thoughts... Using IQFEED DDE I presume you are pulling down the dividend amounts and dividend dates for all 30 symbols which is straight forward...

ie.
=IQLink|mmm!'div rate'/4 = '$0.46'
=IQLink|mmm!'div pay date' = '3/12/06'

All the rates line up with indexarbs updated list of dividend rates...
http://www.indexarb.com/dividendAmountDJIA.html

However it is not entirely clear which dividend payments will be credited and which ones will not. It should be those paid between today and the future expiration but it appears as though there is a mismatch between indexarb and iqlink regarding the payout date of the dividends.

I'm still screwing around with the yield curve to get the right liquid instruments since the yield curve we're interested will primarily be between O/N (ie. overnight) and 3M. According to many sources, "deposit rates" are used to construct the front end of the curve which is O/N thru questionably 3M since they are very liquid. From 3M/6M to about 4Y, the EURODOLLAR futures are the desired liquid instruments for evaluating the yield curve. 4Y+ swap rates. For interpolating the current maturity (ie. 44 days for ES), it is matter of getting quotes from the relevant liquid instruments which is proving trickier than I thought... The interpolation between spot and 3M assuming respective rates of % and 5% would be 4% + (44/105)*(5%-4%) as opposed to averaging... y/n?

Any thoughts...

MAK
 
Quote from makosgu: Any thoughts...
More...
Mak you're always two steps ahead. Thanks for the dde links, I wasn't aware iqfeed had these fields available. This afternoon I cross-referenced the symbols between IQ and idxarb and noticed the issues you describe. I plan to monitor the values and make a manual FV adjustment as they change. It's unfortunate that the dates cannot be relied upon but I'll continue to look into alternatives. Regarding rates, I agree your approach is far more precise. I was simply going for sufficiency and planned to revisit interpolation later. When I looked at averaging the front two ED contracts I was getting within 0.1% of the idxarb rate which worked out to approx a half point diff on the FV. After seeing your progress I'm going to start working on the libor interpolation. Funny thing is that after all of this thought to automating the recentering, I can generally tell when an adjustment needs to be made by watching the tool/charts.
 
Status
Not open for further replies.
Back
Top