Post New

22% cagr since 9/2/97, 1.2 beta, .7 sharpe, -79% worst drawdown vs. for Nasdaq100:
11% cagr since 9/2/97, 1.3 beta, .4 sharpe, -80% worst drawdown
start with Nasdaq100 then from those stocks (101 companies at present) use the following to reduce to 10 companies:
ratio 5 day return excluding today/200 day return excluding today, bottom 10;
then reduce further as follows:
ratio(price,MktCap) bottom 7;
then reduce to 5 stocks further:
ratio(5 day return excluding today/float) bottom 5

Trade monthly
CAGR 21.4% if .2 friction assumption, trade monthly
CAGR 21.8% if trade every 2 months
CAGR 22.3% if trade every 3 months, 1.3 beta, .7 sharpe i.e. better risk/reward with same volatility and drawdown, and twice the return. And its worse start week out-performs the Nas100's best start week with a 16% CAGR v. 11%

https://gtr1.net/2013/?s20081231::nas100.a:nenull:...

I’m always a little skeptical of any screen with only 5 picks and backtests that start after the ’08 bear market.

So upping the min price to $3 and doubling all the cutoff’s and testing from 20071015:
nas100.a != null
aprc(1) > 3
ratio(tr(1,5),tr(1,200)) bottom 20
ratio(aprc(1),MktCap) bottom 14
ratio(tr(1,5),float.s) bottom 10

Brings it down a notch but still does very nicely even with 0.35% trade friction.
CAGR 16.4
SAWR 8.0
MDD -51% high but better than market.
AT 4.1 not that bad

One step further to a final hold 15 CAGR still holding at 15.75%

And then trading only every 3 months CAGR 16.3% perpetual SAWR 9.3%

No longer skeptical!

RAM

start with Nasdaq100 then from those stocks (101 companies at present) use the following to reduce to 10 companies:
ratio 5 day return excluding today/200 day return excluding today, bottom 10;
then reduce further as follows:
ratio(price,MktCap) bottom 7;
then reduce to 5 stocks further:
ratio(5 day return excluding today/float) bottom 5

I can't figure out why this screen works - what about the screening process makes sense, business wise. For example, using ROE is a known positive about company operations success.

Musselmant, what was the thought process about how you came upon this screening process?

Tails

ratio 5 day return excluding today/200 day return excluding today, bottom 10;
then reduce further as follows:
ratio(price,MktCap) bottom 7;
then reduce to 5 stocks further:
ratio(5 day return excluding today/float) bottom 5

Translated to English, that's:

10 stocks with worst 2-through-6 day return compared to 1-201 day return
then of those 10, the 7 stocks with lowest share price compared to company market cap
then of those 7, the 5 stocks with the worst 2-through-6 day return compared to how many $ of company stock are available for trading.

Did I get that right? If so, I don't get how that has anything to do with what the stock might do in the future. It certainly has nothing to do with company metrics. Is it simply a possible reversion to the mean? Does it survive a Mound of Toast test, i.e. 10,20,30 picks; 2,3,6,12 month holds?

Tails

I can't figure out why this screen works

My 2 cents thoughts on why it works. It isn’t screening for the normal financial factors it’s looking for the characteristics of the large cap companies that are swallowing their smaller competitors.
This week picks META, NVIDIA, PDD, Broadcom, Netflix, Palto Alto NetWorks, Crowdstrike are not that different from the ROE_Cash which has a similar bias for total Cash less LTDebt which favors the larger companies. GOOGL, META, TSLA, MSFT, CISCO, COST

Nytimes, Bloomberg and others have also commented on this trend.
Big Companies Are Starting to Swallow the World - The New York Times (nytimes.com)
The Gap Between Large and Small Companies Is Growing. Why? (hbr.org)
Why Do the Biggest Companies Keep Getting Bigger? It’s How They Spend on Tech - WSJ

RAM

With lots of screens I use bad 1 week return sorts and often stocks doing well in last 10 months but with a poor last week is a good combination. I also experiment with float instead of Market Cap since of course actually tradable shares is what matters not total Market Cap. Using just float and dropping MktCap on this one and a $10 minimum with a 3 month hold actually improves the CAGR to 23.96% since 19970902.

https://gtr1.net/2013/?h60::nas100.a:nenull:aprc%2...

If you drop float and just use dol vol instead CAGR 26.97% since 19850201 when the Nasdaq100 was launched.
(29% since end of 2008).

https://gtr1.net/2013/?h60::nas100.a:nenull:aprc%2...

26% cagr with a .3 spread, .75 sharpe, 1.4 beta, no start week worse than 20%, no start week worse sharpe than .6 or worse beta than 1.48.

Low prices and bad last weeks are good on many universes.

10 years ago the average relative bid/ask spread of all additions to the Nasdaq100
since 1998 was 0.33% before the addition, and 0.21% after, so using .3 is plenty.

TY - JOUR
AU - Yu, Susana
AU - Webb, Gwendolyn
AU - Tandon, Kishore
PY - 2014/08/24
SP -
T1 - What Happens When a Stock Is Added to the Nasdaq-100 Index? What Doesn't Happen?
VL - 41
DO - 10.1108/MF-02-2014-0044
JO - Managerial Finance
ER -

Sounds pretty good to me.

As far as I can see, the last two steps don't accomplish anything.

Just pick the five stocks with the worst short-term slump vs long-term return. You probably end up with a lot of high-fliers that just had a bad week, yes? Ulcer Index is 25, for those whose stomach linings are not yet composed entirely of scar tissue...

https://gtr1.net/2013/?nas100.a:nenull:aprc%281%29...

Baltassar

Here's a similar sort of 5-stock thrill ride with a version of BCC that gets you out at when NHNL is bearish:

https://gtr1.net/2013/?~NASDAQ%20top%205%20w%2f%20...

Baltassar

...although the link musselman posted started 20081231, changing the start date to 19970902 as in the post confirms the result... which is over 2x the CAGR of QQQ...

...doubling the cutoffs makes little difference over the longer period, but does drop the CAGR down to ~18 since 10/15/07... guessing, the $2 to $3 price cutoff took out a couple of low-priced explosions.

For comparison since 10/15/07, CAGR on the QQQ itself is 14.7 / LDDD3 12.2 / SAWR 9.4.

Indicating the Naz100 did much better from 10/07 to now than from 9/97 to 10/07 -megacap tech.

ratio(price,MktCap) bottom 7

Am I missing something?
MktCap = price * shares.
Therefor (price / MktCap) = 1 / shares

So ratio(price, MktCap) bottom 7 = shares outstanding top 7.

So that filter is just looking for companies w/ lots of outstanding shares.


John

So that filter is just looking for companies w/ lots of outstanding shares.

Yes, that's correct. It's looking for highly liquid companies, as measured by available outstanding shares. This is also known as the company's "float".

(Floating stock refers to the number of shares a company has available to trade in the open market.)

musselmant also described an S&P 500 version of this screen, with similar rules, which is being tracked as the Liquid Consolidator's screen at https://discussion.fool.com/t/curated-screens-2024...

His original post on the S&P 500 version was at this thread - https://discussion.fool.com/t/5-stock-strategy-w-1...

So that filter is just looking for companies w/ lots of outstanding shares.

Yes, that's correct. It's looking for highly liquid companies, as measured by available outstanding shares. This is also known as the company's "float".


Shares outstanding is not float. This proposed screen also looks for high float in its last step. But the intermediate step that looks for a high share count is senseless. Did AMZN, for example, become a 20 times better investment when it did a 20 to 1 stock split a couple of years ago?

Elan

27% in 2023 and
43% annualized so far in 2024
for the S&P version

Shares outstanding is not float.

I meant that float is shares outstanding that can be traded publicly. Restricted stock, of course, is not available to the public.

"The term float ... is derived by taking a company's outstanding shares and subtracting any restricted stock"

https://www.investopedia.com/ask/answers/what-is-c...

Post New

Berkshire Hathaway FAQ

---

Contact Shrewd'm
Contact the developer of these message boards.