Results for Tu, Sep 3:
[ATTACH=CONFIG]19785[/ATTACH]
Only 3 names were +95 RS Wed, Sep 4 and they are in blue at the top, i.e. Z, BBY, NFLX
Shawn
Printable View
Results for Tu, Sep 3:
[ATTACH=CONFIG]19785[/ATTACH]
Only 3 names were +95 RS Wed, Sep 4 and they are in blue at the top, i.e. Z, BBY, NFLX
Shawn
Hi-
[ATTACH=CONFIG]19800[/ATTACH]
Shawn
[ATTACH=CONFIG]19832[/ATTACH]
Shawn
Hi-
[ATTACH=CONFIG]19853[/ATTACH]
Shawn
Hi:
For Tu, Sep 10 the results were:
[ATTACH=CONFIG]19871[/ATTACH]
Shawn
Hi:
Results for Wed, Sep 11, 2013:
[ATTACH=CONFIG]19886[/ATTACH]
Shawn
Hi-
Results for Th, Sep 12, 2013:
[ATTACH=CONFIG]19908[/ATTACH]
Shawn
Hi
For Fri, Sep 13 the results are:
[ATTACH=CONFIG]19910[/ATTACH]
Shawn
Hi-
I have noticed a more bearish character in the LEV Leaders since about the middle of last week, the overall market seems to be holding up pretty well, so maybe its just a rotation and/or leadership rest.
Results for Mon, Sep 16, 2013:
[ATTACH=CONFIG]19937[/ATTACH]
Shawn
Hi:
Results for Tu, Sep 17:
[ATTACH=CONFIG]19948[/ATTACH]
Shawn
Hi-
Results for Wed, Sep 18:
[ATTACH=CONFIG]19965[/ATTACH]
I am not trading these names mechanically, but it was a surprise and noteworthy that a Chinese coal company made the list ...again..
Shawn
Hi-
Here are the results for Th, Sep 19:
[ATTACH=CONFIG]19977[/ATTACH]
Shawn
Hi-
Here are the results from Fri, Sep 20:
[ATTACH=CONFIG]19982[/ATTACH]
Take care,
Shawn
Hi-
Here are the results for Mon, Sep 23, 2013:
There were no names that made the +95 RS cut-off.. I am not sure what that means.. possibly rotation or a pause?
[ATTACH=CONFIG]20032[/ATTACH]
Take care,
Shawn
Hi-
Here are the results for Tu, Sep 24.
Only 1 name, SWY, made the +95 cut-off
[ATTACH=CONFIG]20053[/ATTACH]
Shawn
Selection is sometimes difficult.
If you are too restrictive, then you end up with too few candidates to be able to operate a good selection.
The fact that there are few long candidates could indicate that funds are not buying leaders for now.
What I observed in the past days however is that the number of shorts candidates is decreasing, which could mean the opposite: funds are not shorting!
I back tested the number of possible candidates related to the average returns but could not find a clear correlation.
What I however found is that when you try to get for example 10 positions, if your selection is too strict, you often end up with being under invested. This means that you miss some of the up/down side.
I hence prefer a larger selection coupled to a system that only crops the five or ten best candidates... or I operate a manual selection, which is my preference, because I like to feel more clever than a computer (in hindsight it is seldom the case.)
Pascal
[QUOTE=Pascal;26499]Selection is sometimes difficult.
If you are too restrictive, then you end up with too few candidates to be able to operate a good selection.
The fact that there are few long candidates could indicate that funds are not buying leaders for now.
What I observed in the past days however is that the number of shorts candidates is decreasing, which could mean the opposite: funds are not shorting!
I back tested the number of possible candidates related to the average returns but could not find a clear correlation.
What I however found is that when you try to get for example 10 positions, if your selection is too strict, you often end up with being under invested. This means that you miss some of the up/down side.
I hence prefer a larger selection coupled to a system that only crops the five or ten best candidates... or I operate a manual selection, which is my preference, because I like to feel more clever than a computer (in hindsight it is seldom the case.)
Pascal[/QUOTE]
I have about 140 stocks to choose from. Choices from several disciplines - value, mechanical and momentum but it sure takes a lot of work !
Trev
Hi-
We see another single name for today BCEI.
The stats being computed here are based on adding a maximum of 4 new names each day near the open, and selling the oldest 4 names near the close. The further assumptions for computing the Avg's, is that each day, 20% of the acct is invested in the new names, no matter what the number of names. This results in a glitch when we have zero names, and we are running only at 80% invested for that particular day, but this has only happened once, so for now I am ignoring that.
Note, that even though this program has only added 2 new positions in the last 3d, there are still 10 names in the portfolio for today. The maximum number of names in the portfolio would be 20, but this would be unusual b/c there are often repeats, like QIHU and TSLA for today.
My sense is that this program/algo ranges between 10-18 names on most day, and that 15-16 names is about the daily avg. This has not been back-tested, etc. and perhaps there are more optimal approaches, but this number of names is reasonable from an implementation perspective and is somewhat diversified from a risk management viewpt.
Results for Wed, Sep 25:
[ATTACH=CONFIG]20058[/ATTACH]
Shawn
Hi-
For Fri, Sep 27:
[ATTACH=CONFIG]20066[/ATTACH]
Have a great weeekend,
Shawn
Hi-
I think I missed this post.
For Th, Sep 26 the results were:
[ATTACH=CONFIG]20067[/ATTACH]
Shawn
Hi-
Here are the monthly and total result since I started May 7. This is a summary of the data published in this thread.
Total returns from close-to-close is +21.4% & open-to-close +15.5% vs SPY close-to-close +4.9%
The end of month (EOM) data is called out in yellow.. note the LEV Leaders generally outperformed SPY each month.
[ATTACH=CONFIG]20069[/ATTACH]
No accommodation is made for market direction, i.e. the algo is always running. I adjusted for the Sep 20 SPY dividend.
Shawn
ps: close-to-close appears to outperform open-to-close, but we will continue to monitor. Note, a rolling 5d portfolio that holds 80% of the portfolio overnight would likely capture most of the close-to-close benefit, if there is one.
Hi-
Results for Mon, Sep 30, 2013:
[ATTACH=CONFIG]20098[/ATTACH]
Shawn
PS: there were 8 unique names in the portfolio for Mon, Sep 30 - this is the lowest I have noted so far..
Hi-
Results for Tu, Oct 1, 2013:
[ATTACH=CONFIG]20124[/ATTACH]
Shawn
Hi-
Results for Wed, Oct 2, 2013:
[ATTACH=CONFIG]20153[/ATTACH]
Shawn
Hi-
Results for Th, Oct 3, 2013
[ATTACH=CONFIG]20196[/ATTACH]
Shawn
Hi-
Results for Fri, Oct 4, 2013
LCI -14% on its secondary announcement was a major drag on performance, but LEV Leaders still finished well above the major ETF benchmarks on Fri.
[ATTACH=CONFIG]20206[/ATTACH]
Take care,
Shawn
Hi-
[ATTACH=CONFIG]20296[/ATTACH]
Take care,
Shawn
Hi-
20d MF is on a Sell Signal.
In the future, I will publish data that assumes that during Sell or Cash signals that this model goes to cash and compare those results to continue to trade these names in adverse market conditions. In this way, we can measure the comparative advantage of sitting out general market downtrends.
The results for Wed, Oct 9, 2013:
[ATTACH=CONFIG]20315[/ATTACH]
Shawn
Hi-
Performance for Th, Oct 10, 2013 (20d MF is on a short signal):
[ATTACH=CONFIG]20336[/ATTACH]
BZ and MOLX were removed b/c they were acquired. I try to avoid merger targets in the LEV Leaders b/c their price most usually do not move up or down much. I missed these 2, but will correct that here.
Shawn
Hi:
Results for Fri, Oct 11, 2013
[ATTACH=CONFIG]20345[/ATTACH]
Shawn
Hi-
For Mon, Oct 14, 2013:
[ATTACH=CONFIG]20350[/ATTACH]
Shawn
Hi,
Results for Tu, Oct 15, 2013
[ATTACH=CONFIG]20354[/ATTACH]
Shawn
Hi-
Results for Wed, Oct 16
[ATTACH=CONFIG]20395[/ATTACH]
Sorry, for the delay.
Shawn
Hi-
Results for Th, Oct 17, 2013
[ATTACH=CONFIG]20396[/ATTACH]
Shawn
Hi-
Results for Fri, Oct 18, 2013:
[ATTACH=CONFIG]20397[/ATTACH]
Shawn
Hi-
Results for Mon, Oct 21, 2013
[ATTACH=CONFIG]20416[/ATTACH]
Shawn
Hi-
For Wed, Oct 23, 2013:
[ATTACH=CONFIG]20454[/ATTACH]
Shawn
Hi-
For Th, Oct 24, 2013:
[ATTACH=CONFIG]20489[/ATTACH]
Shawn
Hi-
Results for Fri, Oct 25, 2013
[ATTACH=CONFIG]20501[/ATTACH]
Take care,
Shawn
Hi-
Results for Mon, Oct 28, 2013
[ATTACH=CONFIG]20544[/ATTACH]
Take care,
Shawn