PDA

View Full Version : Sector Rotation: (if long, XLU; if short, XLF)



Timothy Clontz
05-17-2011, 07:01 AM
My sector rotation model is now complete. This is an adaptation of Len Manskey's model and I do not have permission to give a detailed description of its internals.

However, I can say that the rotation model measures volume divergences between sectors.

The model is currently limited to 9 sector ETFs, but I plan a second model for 39 sector ETFs in the future.

I will personally be using this model in conjunction with the IWM Robot, and am sharing in case anyone else wishes to do the same.

Please note that I will NOT post every day, but I will post on any day that the sector selections change. If there is no new post, that will indicate that the sector selections are unchanged.

If the Robot is in short mode, the sector I will short is Finance (XLF).
If the Robot is in long mode, the sector I will invest long is Utilities (XLU).

Since the Robot is in short mode, short entry on XLF is 16.23, with stop at 16.32

EB
05-17-2011, 12:11 PM
This is interesting Timothy. Are you looking at just the ETFs or the underlying issues?

Timothy Clontz
05-17-2011, 12:28 PM
Great question!

I'm actually selecting the ETFs based on the volume divergences in their underlying issues, rather than on the ETFs themselves.

grems8544
05-17-2011, 01:53 PM
Tim,

Thanks for the work.

It would be useful if you could produce something giving some background and some tested results with dates, simply for correlation. I personally have a difficult time following a new procedure that is a black box with no descriptions.

If you did post such a paper, please point me towards the document.

Regards,

pgd

Timothy Clontz
05-17-2011, 02:17 PM
8357

I've attached a file with the history of the long and short calls, which can be correlated with the IWM Robot's own calls.

I do not have the prices attached, or the specific returns.

A 100% long only test from June 2003 to the present yields 5433 from an initial 987 on the S&P. I'm not exactly sure of the number for the short only test, but it was somewhere around 882 in one of the iterations and 1180 on another (can't remember which correlated with the current calls, because I was only interested in an edge of any kind on the short side and a primary edge on the long side).

I'll get the actual numbers for the short side when I get a little more time (hopefully today or tomorrow).

The specific method, unfortunately, is propriety, and I do not have permission from the author to share more details than "volume divergence". The paper that this is based on is a 1960s MBA thesis that was not commercially published, and the author has been living off of his own profits for decades without managing anyone else's money.

Wish I could do better (and wish I could take credit), but all I did was adapt his model.

Tim

Andrei
05-17-2011, 02:46 PM
Tim,

And do you still have those targets for the S&P, you kindly shared on VIT?

Timothy Clontz
05-17-2011, 03:29 PM
Andrei,

I've been ignoring my Elder Force model while I worked on the sector rotation model.

But because you asked, I plugged in yesterday's numbers to see what I came up with...

1) The Yield Ratio model is 100% short, with a reward to risk ratio of 0/20. The reason is that the Fed has been clamping down on the 3month note, from .16 to .03 with no positive response from the market.

2) The Elder Force model is also short, but at short term support:

1329.47 yesterday's close

Short term:
1349.30 high
1339.43 pivot (stop)
1317.55 low

Long term:
1341.32 high (reverse to long)
1255.70 pivot (target)
1170.07 low

So, last night the model was short with a stop at 1339.43.

Incidentally, the author of the model my sector rotation model was adapted from is also giving a short call.

Everything seems to be in agreement with the Robot (which always gives me pause... I don't like it when everything agrees...)

Tim

Timothy Clontz
05-18-2011, 07:17 AM
I was looking at the wrong numbers on the Robot yesterday. Although not an ideal day to enter the position, these would be the correct numbers if converted from IWM to XLP.

Andrei
05-18-2011, 12:28 PM
Thanks Tim.

I have yet to go through your explanations on sector rotation model, but does it play out as expected, i.e. your call for 1393.28 on 6/8/2011? This is obviously just a projection, so I don't hold my breath for it to be hit on that date, or even that month.

Timothy Clontz
05-18-2011, 01:04 PM
Andrei,

The only thing I was noting was that the average returns for the present relative strength between sectors is still (at least late stage) bullish.

Right now the average returns would take us to 1439.01 on 6/16. Will we actually hit that?

I doubt it, because this was an artificial bull market propped up by Fed steroids. I'm tempted to call it the "bulloney market".

My favorite short is XLP and my favorite long is XLU. I see that XLP is crushing XLU right now...

Averages are only averages.

My other models are bearish. My yield ratio model is SERIOUSLY bearish. But it's ALSO skewed by the Fed action.

In other words, I don't think ANY average models apply to the present state of the market.

Tim