Index | Recent Threads | Who's Online | Search

Posts: 100    Pages: 10    Prev 1 2 3 4 5 6 7 8 9 10 Next
Last Post
New Thread
This topic has been viewed 6093 times and has 99 replies
judgetrade
Re: Improvement of my Ranking System

By the way I had a good week :-)

I rebalnced last friday (11/08) istead of last monday 11/11 and catched XPEL before the pop.
You got to be lucky.

Best Regards
Andreas

Attachment aaa.png (300897 bytes) (Download count: 243)


Nov 16, 2019 8:38:12 AM       
geov
Re: Improvement of my Ranking System

Hi Andreas,
There is something wrong with the Trade Stats you posted. Must be for another model or different slippage assumption.
The sim shows a Total Market Value= $77,684,603.
The Trade Stats shows a Ending Market Value= $160,770,485.
Those numbers must be the same.

Annualized return from Trade Stats: 51.14% not 46.03% as per sim.
1/2/1999 11/18/2019
-$28,777 $160,770,485 51.14%

Nov 16, 2019 1:05:54 PM       
geov
Re: Improvement of my Ranking System

One can get similar performance from ETFs with a lot less effort and much lower trading costs.

Here is a sim with the same starting capital of $28,777 and backtest period as Andreas' model.
Annualized Return is higher (48%), and max D/D lower.

Positive returns and excess return over SPY for all calendar years.

Attachment Leveraged ETF port.png (166944 bytes) (Download count: 229)


Attachment Leveraged ETF trade stats.png (121498 bytes) (Download count: 228)


Attachment Leveraged ETF perform.png (127691 bytes) (Download count: 230)


Nov 16, 2019 3:00:38 PM       
Edit 3 times, last edit by geov at Nov 16, 2019 3:17:36 PM
WERNER
Re: Improvement of my Ranking System

Hi Georg,
that's impressive, to say the least!
Will you be offering this here as a DM or at iMarketSignals (or a version thereof)?

Werner

Nov 16, 2019 3:32:48 PM       
InspectorSector
Re: Improvement of my Ranking System

Better in recent years, but slightly worse then EPS%ChgPYQ


Andreas - I noticed that you do not have revenue growth as a factor. If it was my RS I would be using Revenue Growth as the primary measure of growth, not earnings. It took the genius of Jeff Bezos to make me realize that earnings are not a particularly good measure of anything. We have been sold a fantasy by all those value investing legends. Why would a company show earnings as they are simply going to be taxed away by the government? Investors don't benefit directly from earnings except to justify dividend payouts which most of us don't care about unless we are investing for income.

I know that Benjamin Graham is going to roll over in his grave and I'll probably take a lot of flack for saying this BUT ... It isn't in the investor's benefit to chase earnings. Amazon, arguably the most successful company ever, only comes up for air (shows earnings) when they want something, like a capital raise for example. Other than that they are not useful and instead, Amazon feeds all profits back into capturing market share and expanding its Total Addressable Market.

Lately, I have been leaning towards using revenue growth as the primary measure. 3-Yr composite growth works well. Then, as additional factors, look for improvements in margin... gross margin, operating margin, etc. Margin improvement RS factors provide additive benefit to the base factor which is revenue growth. The alternative is to provide both revenue growth and earnings growth, factors which do not have additive benefit. If fact the earnings growth is problematic as was stated before and simply muddies the waters when employed in parallel to the "clean" revenue growth factor. In other words:

Good combination of factors:
(1) 3Yr revenue growth +
(2) Gross Margin improvement +
(3) Operating margin improvement

Bad combination of factors:
(1) 3Yr revenue growth +
(2) earnings growth

Steve

Nov 16, 2019 3:36:03 PM       
geov
Re: Improvement of my Ranking System

Werner, not sure that people will want to invest in those ETFs -- a bit risky.
SSO, UPRO, SVXY, DDM, QLD UST, UBT.

But interesting that the D/D is so low with those leveraged ETFs. This may be because it always holds 2 ETFs.
Current holdings: 87% QLD + 13% UST. So there is a bit of protection here.

Nov 16, 2019 4:30:23 PM       
SpacemanJones
Re: Improvement of my Ranking System

Hi Andreas, I haven't been able to successfully work momentum factors into my models and I don't know if this is applicable in your microcap universe, but in most cases I've worked with momentum I've found that short term momentum tends to be mean reverting. I just wanted to suggest that for industry momentum it might be worth dropping/eliminating the 4wk industry price momentum (still use 13wk, 26wk, and 52wk) and see if it makes a difference due to the short term mean reverting tendency in prices on a 4wk timescale.

Nov 16, 2019 7:51:37 PM       
Edit 1 times, last edit by SpacemanJones at Nov 16, 2019 7:52:27 PM
judgetrade
Re: Improvement of my Ranking System

HI SpacemanJones,,

tried it out, gets a little less good backtest... thank you for the hint..

I think the reason why it works is, that it is not only 4 Week momentum but a combination. read some papers, where this was suggested (to combine different momentum look backs and combine short term with longer term...)
Best Regards
Andreas

Nov 18, 2019 2:05:42 AM       
Edit 1 times, last edit by judgetrade at Nov 18, 2019 2:10:52 AM
judgetrade
Re: Improvement of my Ranking System

Hi Steve, thank you for the hint, I will try it out...
Best Regards
Andreas

Nov 18, 2019 2:06:21 AM       
judgetrade
Re: Improvement of my Ranking System

Hi Georg,
yes, this one is with 0.3% Slippage, not variable! the good think is though that the difference between 0.3% and variable is not high (about 4-5% ann.) That is the actual breakthrough to have found a small cap (before that nano cap) model, that performs!

Have to look into the ETF Space, the think is though its then mostly momentum and macro and no fundamentels and I can not capture the size effect as good, so stocks are my thing.

thank you!!!!

Best Regards
Andreas

Nov 18, 2019 2:09:25 AM       
Posts: 100    Pages: 10    Prev 1 2 3 4 5 6 7 8 9 10 Next
 Last Post