Jump to content



Photo

Oil and Oil Services trade update...


  • Please log in to reply
8 replies to this topic

#1 A-ha

A-ha

    Member

  • Traders-Talk User
  • 5,875 posts

Posted 05 January 2007 - 05:57 PM

As a trading vehicle , I am in love with this sector. Lots of volatility, and quite predictable because very institutional. I think the sector as well as the crude will keep going higher early next week. I accumulated some USO at 47.4 and WFT and OIH at 36.5 and 131.6 respectively yesterday as posted real-time. I plan to close USO at 49+, WFT around 40 and OIH around 135 early next week. After that, I expect at least a test of the lows. OIH today touched the exact intermediate term target ($129) I gave 4 weeks ago. Day's low was 128.95 But now I am not confident on its durability over the intermediate term time frame. In other words, after this initial rebound, the sector can completely break down. If it does, the next target comes in play which is $106-$108. I also plan to short the coming bounce but I realize that it is too obvious. So if we get the bounce, I think there may be a few surprises before the sector start moving in the obvious direction again.

Edited by xD&Cox, 05 January 2007 - 06:00 PM.


#2 arbman

arbman

    Quant

  • Traders-Talk User
  • 19,504 posts

Posted 06 January 2007 - 09:55 AM

This is similar to what I am using, I am using a differential method to extract the latest patterns rather than the longer term secular ones, but this one is worth to observe here...

This is the normalized volatility patterns according to the 95% confidence interval (3 standard deviations) during the logarithmic rise of the 2002 - July 2006 period.

Posted Image


The patterns are better extracted during a secular trend, here's the 75% of the pattern intensities during the rise. There is visible right translation in the cycles, I suspect it will change and shift to the left if the secular trend has turned or turning to neutral for now [LINK]. So the above chart might favor more of the waterfall patterns during the declines...

Posted Image


Notice that the trend has indeed changed since the bounce has been considerably weak in the second half of the 2006, but the volatility path favors the upside or stable prices during the first half of the 2007...

I posted the periodicity transform paper here a while back...

- kisa

Edited by kisacik, 06 January 2007 - 10:00 AM.


#3 A-ha

A-ha

    Member

  • Traders-Talk User
  • 5,875 posts

Posted 06 January 2007 - 02:13 PM

I am sure PT is not widely used in finance yet and I am very confident about its success over Fourier Transform which has been mindlessly used by cycle crowd to determine fundamental harmonics which are orthogonal ! :lol: I had written a paper on pattern recognition utilizing KLT transform when I was pursuing PhD. In a sense, PT is an opposite of KLT transform which is the optimal orthogonal transform built on orthogonal eigen components. PT is built on periodic components. Do you have a Matlab code for PT?

#4 arbman

arbman

    Quant

  • Traders-Talk User
  • 19,504 posts

Posted 06 January 2007 - 07:59 PM

The matlab code is here.

#5 arbman

arbman

    Quant

  • Traders-Talk User
  • 19,504 posts

Posted 06 January 2007 - 08:31 PM

I must tell you, unlike principal component analysis (KLT) or any other orthogonal method, PT requires an understanding of what is important in the data and its decomposition. There is no unique solution, I favor maximizing the signal to noise ratio by using my own metrics, but I do not decompose the entire data in one pass or large matrix. My methods are iterative and it will penalize and rebalance. But, I do decompose into the likely patterns that are not orthogonal too, but the real strength of my method comes from the ability of the evaluation of the likely momentum and the advection within the likely periodicities, or patterns, in multiple time frames. I actually do not assume any basis functions or patterns strictly either...

- kisa

Edited by kisacik, 06 January 2007 - 08:33 PM.


#6 A-ha

A-ha

    Member

  • Traders-Talk User
  • 5,875 posts

Posted 07 January 2007 - 10:37 AM

I must tell you, unlike principal component analysis (KLT) or any other orthogonal method, PT requires an understanding of what is important in the data and its decomposition. There is no unique solution, I favor maximizing the signal to noise ratio by using my own metrics, but I do not decompose the entire data in one pass or large matrix. My methods are iterative and it will penalize and rebalance. But, I do decompose into the likely patterns that are not orthogonal too, but the real strength of my method comes from the ability of the evaluation of the likely momentum and the advection within the likely periodicities, or patterns, in multiple time frames. I actually do not assume any basis functions or patterns strictly either...

- kisa


So basically you take several PTs of different time frames using different parameters and you measure the likelihood of each by its moments? You say momentum, I dont know what is momentum in this contex but if it is the moment descriptives you talking about then you should use several of them to identify a pattern. Moments are descriptive quantities that can be used to represent patterns to some extend. But you need at least 7 or more moments for each pattern. Also there are different type of moments, Hu's moments could give better results than standardized moments.
I assume you use a scripting language to compute. You would go nuts with C++ or other low level tools.

#7 arbman

arbman

    Quant

  • Traders-Talk User
  • 19,504 posts

Posted 07 January 2007 - 02:54 PM

I encapsulated the core C++ modules for python, yes it is kind of many PTs. I think there is generally little use for any global decomposition method, instead I focused on finding many local and the transient patterns by higher order moments. I took a bunch of shortcuts though and I am more interested in the local "momentum" than the "pattern" in the sense that the distorsions in the existing patterns come from the transient patterns, not necessarily a repetetive pattern that is...

#8 arbman

arbman

    Quant

  • Traders-Talk User
  • 19,504 posts

Posted 08 January 2007 - 02:55 AM

Since this discussion has been taking place here and I am sure we look like quite from Mars to the many, I wanted to point out to this paper to raise more interest; transient signal detection using higher order statistics (moments)...

#9 slatedrake

slatedrake

    Member

  • Traders-Talk User
  • 269 posts

Posted 08 January 2007 - 10:47 PM

Hey Xd, Just wondering if you decided to close these out this morning or still holding. TIA.
Before you start trading get your brain around risk control. Know how much leverage you're using and know when to go to cash if you're wrong.