PDA

View Full Version : Sector Selection / IWM comparison, part 5



Timothy Clontz
08-02-2011, 12:10 PM
Part 1, without Billy's pivot help:
IWM 6.51%
XLU 0.91%
Difference -5.60%
Cumulative -5.60%

Part 2, with Billy's pivot help:
IWM -2.21%
XLK 0.00% (i.e. the trade never hit the stop loss and remained active)
Difference 2.21%
Cumulative -3.39%

Part 3, with Billy's pivot help:
IWM -1.79%
XLK 1.31%
Difference 3.10%
Cumulative -0.29%

Part 4, with Billy's pivot help:
IWM -2.11%
XLK 0.00% (i.e. the trade never hit the stop loss and remained active)
Difference 2.11%
Cumulative 1.82%

Part 5, with Billy's pivot help:
IWM -2.39%
XLK -2.00%
Difference 0.39%
Cumulative 2.21%

Pivot only cumulative (Part 2-5) 7.81%

***NOTE: updated the results as XLK got stopped out as well.

Once again, as shown in the difference between the first trade and the others, the test could not be conducted without Billy's pivots.

I'd also like to note that there is no way ANY model could reliably operate under unprecedented conditions. We have had, for the first time in history, a President actively trying to spook the markets. It was only a few days ago when his spokesman scowled into the camera and threatened, "Buy and hold; see what happens." This is completely outside of the parameters of any mechanical algorithm, and one that WOULD operate correctly here would quite likely not work any other time.

Given that, my test is still in the first of four stages (two long and two short sequences). We are still in the first long sequence, and comparison of my sector adaptation to the IWM baseline is still early. The relative performance at the moment could still just be odd luck.

While I'm happy to be the guinea pig, it's still early in a live beta test.

Tim

Timothy Clontz
08-03-2011, 07:11 AM
The long sector is XLK.

Had the Robot gone short the short sector would have been XLE.

I am attempting to enter with the IWM Robot at the following settings:

XLK long 24.96, with stop at 24.41

***NOTE: the XLK buy limit executed at 24.96.

Timothy Clontz
08-04-2011, 06:31 AM
Kudos to Billy for his pivot entry points. XLK bottomed at only 4 cents below the pivot entry.

In spite of the unusual behavior of IWM at present, I'm extremely sold on the model developed by Pascal and Billy. Part of the problem we've been experiencing (and I'm trading IWM too) could be related to sector rotation. IWM is the entire market, but within that market sectors reshuffle their relative strength and money flow at major trend transitions. This could cause unpredictable volatility changes for the market as a whole.

Think of IWM as the womb, with the internal sectors like twins thrashing with all eight of their combined hands and feet. While the woman herself may not give any tells she's likely to jump, the twins might make her move in a way not even she can expect.

XLK entered the trade at 24.96, and today the stop appears to be 24.90 (which curiously is just below yesterday's low).

A hypothetical new entry for today would be a long buy at 25.36, with a stop at 24.76.

Just a note about some things I am watching for in this test:

Billy and I had a discussion a few weeks ago about the total volatility of IWM as it relates to the market. IWM is a small cap value index, while my sector ETFs are large cap weighted. It's quite possible that IWM could outperform even the best sector ETF from the simple beta difference between large caps and small caps. Also, my sector model was designed to be long-only, while the Robot alternates between both short and long positions. While I AM able to determine the sector to choose for a short trade on my model, I do not have enough data to back test it on the Robot timing.

Finally, it is entirely possible for two models that work well separately to actually negate each other's benefit when combined together. There's just no way to know without testing it live.

It would therefore be premature to make anything of the slight edge the sectors are adding at the moment. It could be odd luck from lower beta on an atypical move down that won't work in more typical conditions.

In other words, right now it's still in the realm of odd luck.

Tim

Timothy Clontz
08-04-2011, 09:35 AM
Buy on IWM hit at 76.05; I've adjusted the stop to 73.82 accordingly.

Billy, please let me know if this is incorrect. I'm trying to follow your IWM model exactly per your ruleset.

Tim

adam ali
08-04-2011, 09:45 AM
Just to reconfirm the rules, even though the 20DMF is in cash, the Robot is still looking for an entry - long or short. Given the price setup available today and associated rules of the pivot points, the Robot will determine whether there is a 3:1 reward to risk opportunity, either long or short. In the case of today, there was no "edge" on the short side, but there is one on the long side. That's why the Robot is long.

Sorry for laboring over a point that's been well documented here. Do I have this right?

Timothy Clontz
08-04-2011, 09:47 AM
I wasn't questioning WHETHER to enter, but exactly what price to adjust the stop to. The stop is a certain percentage below the entry, and should be lowered based on your actual entry point. I just wanted to know if I calculated it right.

That might also affect the precise stops I place on XLK. I think I might be a few cents off.

Timothy Clontz
08-04-2011, 11:06 AM
XLK hit the stop loss and exited for a 0.20% loss. Will update in a new thread.

Billy
08-04-2011, 11:30 AM
Buy on IWM hit at 76.05; I've adjusted the stop to 73.82 accordingly.

Billy, please let me know if this is incorrect. I'm trying to follow your IWM model exactly per your ruleset.

Tim

Sorry Tim, I missed your post.
Yes, you are correct, the stop today on IWM is 2.925% below entry.
Billy

Timothy Clontz
08-04-2011, 11:30 AM
Quite surprising the sheer intensity of this meltdown.