Skip to main content
Message Font: Serif | Sans-Serif
 
No. of Recommendations: 8
Here's a Zacks screen that has done pretty well based very
loosely on the "surfing" posts on MI and Jim Cramer's recent
shows on finding stocks that have a dividend rate of >4%.

Screen-
Dividend yield>4%
52 week return top 25%
12 week return top 5

Below is some output from Zacks
note the CAGR of 54.3% with a drawdown of -14.5%.
Below that is the the actual results on how the
$10K went to $332K.

If you add Zacks=1 to the above 3 criteria, the
CAGR will jump up to 59.7% with a drawdown at -10.8%
(results at bottom of post)

All results used a 12 week hold period.

STATISTICS  ex.: $10,000 start       Strategy  S&P 500
Total Compounded Return 3222.6 9.8
Total Compounded Return $ $332,260 $10,979
Compounded Annual Growth Rate 54.3 1.2
Win Ratio 69 57
Winning Periods/Total Periods 24 of 35 20 of 35
Avg. # of Stocks Held 5
Avg. Periodic Turnover 92.9
Avg. Return per Period 11.7 0.5
Avg. Winning Period 19.3 5
Largest Winning Period 68.8 16.7
Avg. Losing Period -4.8 -5.5
Largest Losing Period -13 -14.2
Max. Drawdown -14.5 -44.1
Avg. Winning Stretch (# of Periods) 3 2.9
Best Stretch (# of Periods) 6 6
Avg. Losing Stretch (# of Periods) 1.4 2
Worst Stretch (# of Periods) 2 5


Period          Date        Stocks  Return  S&P    Diff         $10,000
1 12/17/1999 5 68.8 -1.8 70.6 16,880
2 3/10/2000 5 -6.4 5.9 -12.3 15,800
3 6/2/2000 5 15.9 2.0 13.9 18,312
4 8/25/2000 5 -2.6 -9.2 6.6 17,836
5 11/17/2000 5 20.9 -3.9 24.8 21,563
6 2/9/2001 5 19.0 -3.7 22.7 25,660
7 5/4/2001 5 22.3 -4.8 27.1 31,383
8 7/27/2001 5 -2.7 -11.0 8.3 30,535
9 10/19/2001 5 -4.1 6.7 -10.8 29,283
10 1/11/2002 5 43.9 -2.0 45.9 42,139
11 4/5/2002 5 -1.7 -11.5 9.8 41,422
12 6/28/2002 5 -13.0 -14.2 1.2 36,038
13 9/20/2002 5 12.1 5.7 6.4 40,398
14 12/13/2002 5 2.6 -6.4 9.0 41,448
15 3/7/2003 5 54.8 16.7 38.0 64,162
16 5/30/2003 5 5.6 3.4 2.1 67,755
17 8/22/2003 5 10.6 6.1 4.5 74,937
18 11/14/2003 5 21.7 9.2 12.5 91,199
19 2/6/2004 5 -8.7 -2.7 -6.0 83,264
20 4/30/2004 5 2.9 -1.5 4.4 85,679
21 7/23/2004 5 16.4 2.4 14.0 99,731
22 10/15/2004 5 11.3 7.4 3.9 111,000
23 1/7/2005 5 15.7 -0.7 16.4 128,427
24 4/1/2005 5 6.3 2.0 4.3 136,518
25 6/24/2005 5 -0.2 4.3 -4.5 136,245
26 9/16/2005 5 -1.3 2.1 -3.4 134,474
27 12/9/2005 5 20.3 2.6 17.7 161,772
28 3/3/2006 5 -8.7 -0.1 -8.6 147,698
29 5/26/2006 5 10.1 2.2 7.9 162,615
30 8/18/2006 5 9.3 6.5 2.8 177,738
31 11/10/2006 5 30.4 5.3 25.1 231,771
32 2/2/2007 5 18.1 3.6 14.6 273,722
33 4/27/2007 5 7.5 3.1 4.4 294,251
34 7/20/2007 5 17.5 2.2 15.3 345,745
35 10/12/2007 5 -3.9 -9.2 5.3 332,260

Average 5 11.7 0.5 11.3

Up Markets 20 5 12.8 5.0 7.8

Down Markets 15 5 10.3 -5.5 15.8

Annualized 61.7 2.1 58.7

--------------------------------------------------------------

Adding Zacks=1 to the criteria gives a very nice screen-

STATISTICS  ex.: $10,000 start       Strategy  S&P 500
Total Compounded Return 4295.7 9.8
Total Compounded Return $ $439,566 $10,979
Compounded Annual Growth Rate 59.7 1.2
Win Ratio 83 57
Winning Periods/Total Periods 29 of 35 20 of 35
Avg. # of Stocks Held 3.2
Avg. Periodic Turnover 87.7
Avg. Return per Period 12.5 0.5
Avg. Winning Period 16.3 5
Largest Winning Period 87.4 16.7
Avg. Losing Period -6.2 -5.5
Largest Losing Period -10 -14.2
Max. Drawdown -10.8 -44.1
Avg. Winning Stretch (# of Periods) 5.8 2.9
Best Stretch (# of Periods) 11 6
Avg. Losing Stretch (# of Periods) 1.3 2
Worst Stretch (# of Periods) 2 5


Period          Date        Comp.  Return  S&P    Diff    $     10,000
1 12/17/1999 5 7.0 -1.8 8.8 10,700
2 3/10/2000 5 9.9 5.9 4.0 11,759
3 6/2/2000 5 14.0 2.0 12.0 13,406
4 8/25/2000 5 6.1 -9.2 15.3 14,223
5 11/17/2000 5 5.6 -3.9 9.5 15,020
6 2/9/2001 5 10.1 -3.7 13.8 16,537
7 5/4/2001 5 22.9 -4.8 27.7 20,324
8 7/27/2001 5 3.4 -11.0 14.4 21,015
9 10/19/2001 5 6.1 6.7 -0.6 22,297
10 1/11/2002 5 21.6 -2.0 23.6 27,113
11 4/5/2002 1 87.4 -11.5 98.9 50,809
12 6/28/2002 3 -9.4 -14.2 4.8 46,033
13 9/20/2002 3 -1.6 5.7 -7.3 45,297
14 12/13/2002 1 29.1 -6.4 35.5 58,478
15 3/7/2003 3 31.8 16.7 15.1 77,074
16 5/30/2003 2 10.4 3.4 6.9 85,090
17 8/22/2003 2 22.8 6.1 16.7 104,490
18 11/14/2003 2 8.7 9.2 -0.4 113,581
19 2/6/2004 1 9.2 -2.7 11.9 124,030
20 4/30/2004 2 16.9 -1.5 18.5 144,992
21 7/23/2004 1 26.1 2.4 23.7 182,834
22 10/15/2004 3 8.0 7.4 0.6 197,461
23 1/7/2005 4 11.6 -0.7 12.3 220,367
24 4/1/2005 2 -10.0 2.0 -12.0 198,330
25 6/24/2005 2 11.7 4.3 7.4 221,535
26 9/16/2005 2 -9.2 2.1 -11.4 201,153
27 12/9/2005 1 25.7 2.6 23.1 252,850
28 3/3/2006 3 -0.2 -0.1 0.0 252,344
29 5/26/2006 3 2.0 2.2 -0.2 257,391
30 8/18/2006 3 13.7 6.5 7.3 292,654
31 11/10/2006 3 17.7 5.3 12.4 344,453
32 2/2/2007 3 16.0 3.6 12.4 399,566
33 4/27/2007 3 3.3 3.1 0.2 412,752
34 7/20/2007 3 13.9 2.2 11.7 470,124
35 10/12/2007 5 -6.5 -9.2 2.7 439,566
Average
3.2 12.5 0.5 12.0
Up Markets
20 2.8 11.0 5.0 6.1
Down Markets
15 3.7 14.3 -5.5 19.8
Annualized
66.3 2.1 63.3
Print the post Back To Top
No. of Recommendations: 0
Is this doable with other databases such as MSN? Returns?
Print the post Back To Top
No. of Recommendations: 0
ep0001-
I made up this screen in MSN but it gives a lot of ADRs(which might be good!).
So I added the exchange to be NYSE.

Alas, backtesting is not possible in MSN's screener.
Print the post Back To Top
No. of Recommendations: 0
Are you implying that you can't do the backtesting using Zack's software for the exact screen you produced with MSN?
Print the post Back To Top
No. of Recommendations: 1
nivra0-
I'm not sure that all the foreign stocks (ADRs, etc.) that
are in MSN are in Zacks.
If you run "simple" MI/VL screen, do you think you will
get the same results as from a MSN or Zacks screen.
Doubtful. They might be close, but not exact on some
of the criteria used.
Also defintions are somewhat differing- Zacks has a
24 week price increase variable (twice as long as quarter), VL a 26 week one (half of a 52 week)
Print the post Back To Top
No. of Recommendations: 2
A quick check using Jamie Gritton's backtester gives a nice result, although less spectacular than the Zacks' backtester screen:

http://backtest.org/9708h3SBT15XdyldG4Xtr1yT25HXr13T5

Timeliness™ 1-5; Div'd Yield > 4; Total Return 1-Year top 25%; Total Return 13-Week top 5 . 3 months hold, starting Jan 1997 till today (?)

Screen S&P

CAGR 13 7
GSD 15 15
Sharpe Ratio 0.70 0.29

The maximum DD is 21%

The all months summary is even better, giving an average CAGR =16

The dividend yield field is available only starting from 1997. However a "current dividend" (quarterly? annual??) field is available since 1986.

Replacing dividend yield with (current dividend)/price allows to extend the backtest to 1986 and the results are pretty much the same:

http://backtest.org/8608h3SBT15XOcdvDpriCG.04Xtr1yT25HXr13T5

Timeliness™ 1-5; (Current Dividend / Stock Price) > .04; Total Return 1-Year top 25%; Total Return 13-Week top 5 . 3 months hold starting Jan 1986 till today (?)

Screen S&P

CAGR 14 11
GSD 17 16
Sharpe Ratio 0.65 0.50

The maximum DD is about 30%

I like the simplicity of the screen criteria. It would be also nice to backtest it on other markets: I am a european investor and I am curious to know what the result would be on British or German securities. Does anybody have access to a backtester for European markets?

eulero
Print the post Back To Top
No. of Recommendations: 0
One last ?'s, maybe.

How do you do 52 week return top 25% in MSN?

52 week return / P/E top 25 %

Thanks
Print the post Back To Top
No. of Recommendations: 0
My approach to run this on MSN is as follows:

% Price change last year/high as possible
% Price change last qtr/display only
When you get the screen, click on the heading % Pice change last qtr and you can get the top five (descending).

This approach is somewhat different because it does not get the top 25% but I don't know how to do that.
Print the post Back To Top
No. of Recommendations: 0
Yes, in addition to pbd200's logic, when you
add that dividend >4% and exchange = NYSE
you come pretty close to the Zacks list.
The only thing missing is if you want to use
the Zacks proprietary rating (but MSN has their own rating
if you want to give that a try).

One might want to check out the payout ratio (div/earnings)
before plunking down the ol' $$$$$.

Making the list, I see that WWE (World Wrestling Entertainment) is paying close to 8%.
I might buy that just to get the annual report.
Has market cap. over $1B, gets a strong buy from Zacks
and an 8 (10 max) from MSN.
Print the post Back To Top
No. of Recommendations: 0
To check if I am doing this correctly,

list mostly royalty trusts?

If div > earnings, should avoid?

Did I miss the std. dev.?
Print the post Back To Top
No. of Recommendations: 1
Are the Zacks screens posted by anyone?
Print the post Back To Top
No. of Recommendations: 0
Are the Zacks screens posted by anyone? <iB>


I am going to assume not. In which case, to confirm that my results using MoneyCentral are in the ball park; does this screen give a lot of royalty trusts?
Print the post Back To Top
No. of Recommendations: 0
Current picks using MSN are:

GCT
PDS
PBT
FDG
TNE
Print the post Back To Top
No. of Recommendations: 0
I get the following using NYSE and >4% div.
FDG
PBT
TNE
BPT
ASA

I don't get GCT and PDS. The past 52 weeks for GCT was negative 11.9%
and for PDS it was negative 1.8%.

Also, even for my list I would look at their graphs and not buy
TNE or ASA because today they are below their 50-day moving average.
Print the post Back To Top
No. of Recommendations: 0
The past 52 weeks for GCT was negative 11.9%

Interesting. Why would MSN show a positive 52 wk relative strength when the chart shows a decline?
Print the post Back To Top
No. of Recommendations: 0
This approach is somewhat different because it does not get the top 25% but I don't know how to do that.


One suggestion. Go to Zack's free screener and determine that there are 660 +/= stocks paying > 4% and on NYSE.

Then use MSN and limit to 165 stocks.

someone also commented using the MSN rater to simulate Zacks. I don't know if there is an equivalency. A lot of the stocks selected are not rated by MSN
Print the post Back To Top
No. of Recommendations: 0
My approach to run this on MSN is as follows:

% Price change last year/high as possible
% Price change last qtr/display only
When you get the screen, click on the heading % Pice change last qtr and you can get the top five (descending).

This approach is somewhat different because it does not get the top 25% but I don't know how to do that.


A thought, Run the screen as you describe. Determine the number of stocks, for example 100. Then adjust the % Price change last years to be >= ? to give 25 stocks (top 25%), great.
Print the post Back To Top
No. of Recommendations: 0
Are there any results from this screen as a monthly, bi-monthly?
Print the post Back To Top
No. of Recommendations: 2
One more ? How many stocks pass the screen for 4% div, top 25 % of 52 wk return and zacks = #1?
Print the post Back To Top
No. of Recommendations: 0
>4% div
z=1
top 25% 52 wk return

avg 9 stocks pass
cagr 25%
max drawdown 25%
Print the post Back To Top
No. of Recommendations: 1
The return drops in half if you add price>$1 and (price x ave 20 day volume >$100,000) liquidty to the test, with or without zacks rank=1. And for the post-discovery period it gives 0% without zacks rank and 7% with it, averaged over all possible starts.

The good reported returns appear to be generated by illiquid stocks. Also, with the zacks rank you often don't get 5 stocks picked with a reasonable liquidity requirement.
Print the post Back To Top