StockFetcher Forums · Filter Exchange · Interesting filter - perhaps the best I've seen so far<< 1 ... 3 4 5 6 7 ... 8 >>Post Follow-up
Kevin_in_GA
4,557 posts
msg #78512
Ignore Kevin_in_GA
modified
9/2/2009 4:08:22 PM

Here's where I see an issue - your trading costs on those 99 trades would be, conservatively $8 for the buy/sell combo at Sharebuilder. That means that you paid ~$800 in commissions, and then you have to pay taxes on your gains. Since your equity increase was only a hair over $1000, you made, on average, $2 per trade before taxes.

You net out negative because you took small profits rather than letting them run for a bit longer. My run on the same stock and timing generated the following:

There were 89 total stocks entered. Of those, 39 or 43.82% were complete and 50 or 56.18% were open.
Of the 39 completed trades, 38 trades or 97.44%resulted in a net gain.
Your average net change for completed trades was: 22.79%.
The average draw down of your approach was: -7.41%.
The average max profit of your approach was: 27.07%
The Reward/Risk ratio for this approach is: 168.69
Annualized Return on Investment (ROI): 156.36%, the ROI of ^SPX was: 41.72%.

Total (Cash + Market): $124122.01

Note that under these conditions you make 24 times the equity on fewer trades.

Would have been higher at the end if we hadn't tanked in the market yesterday because of the 50 open positions which are closed at the end of the time window.



Kevin_in_GA
4,557 posts
msg #78513
Ignore Kevin_in_GA
9/2/2009 4:20:59 PM

also which of the stocks today are you planning on buying? thanks, sam

++++++++++++++++++++++++++++++++++++++++++++++++++++++

I won't know until the end of the day. I get the results emailed to me each night, and look at them then.

However, running the screen right now it looks like:

ACAD
AFSI
ARTG
CA
CACC
DLB
FHCO
GMTN
GVA
NCIT
NRG
SAN
SLF


I keep seeing ACAD, NCIT, and GMTN over the last few days (I have not placed any trades on this system since last week). That concerns me because it indicates that I could pick these up cheaper by waiting. Perhaps I need to include a confirmation condition, like

Fetcher[
close 1 day ago was below lower bollinger bands(16,2.5) 1 day ago
and close is above close 1 day ago
and close is above 1 and average volume(50) is above 50000
]



this will find stocks that closed below the lower BB yesterday, but bounced up today - could also use close is above lower BB as well.

Kevin

Kevin_in_GA
4,557 posts
msg #78517
Ignore Kevin_in_GA
9/2/2009 7:07:25 PM

Just for fun, here is a tweaked version:

Test started on 12/31/2008 ended on 05/01/2009, covering 83 days
Filter used:

Fetcher[/*NEW BOLLINGER BAND BUY FILTER SEPT 2ND 2009*/

CLOSE 1 DAY AGO IS BELOW LOWER BOLLINGER BANDS(16,2.5) 1 DAY AGO
AND CLOSE IS ABOVE LOWER BOLLINGER BANDS(16,2.5)
AND VOLUME IS ABOVE 50000
and apply to watchlist(IBD_TOP_180,IBD2,IBD3)
]



Trade Statistics
There were 78 total stocks entered. Of those, 40 or 51.28% were complete and 38 or 48.72% were open.
Of the 40 completed trades, 34 trades or 85.00%resulted in a net gain.
Your average net change for completed trades was: 22.24%.
The average draw down of your approach was: -19.16%.
The average max profit of your approach was: 27.06%
The Reward/Risk ratio for this approach is: 13.43
Annualized Return on Investment (ROI): 168.54%, the ROI of ^SPX was: -4.37%.

EQUITY SUMMARY: Total (Cash + Market): $128596.73



Test started on 05/01/2009 ended on 09/01/2009, covering 85 days

Trade Statistics
There were 87 total stocks entered. Of those, 37 or 42.53% were complete and 50 or 57.47% were open.
Of the 37 completed trades, 35 trades or 94.59%resulted in a net gain.
Your average net change for completed trades was: 22.42%.
The average draw down of your approach was: -6.93%.
The average max profit of your approach was: 25.87%
The Reward/Risk ratio for this approach is: 179.78
Annualized Return on Investment (ROI): 181.32%, the ROI of ^SPX was: 41.72%.

EQUITY SUMMARY: Total (Cash + Market): $125166.43


trendscanner
265 posts
msg #78518
Ignore trendscanner
9/2/2009 8:21:21 PM

Kevin, thanks for posting the lists for the IBD stocks and the BB scans as well. The BB scan you've developed for the IBD stocks is simply brilliant and brilliantly simple. It's easily implemented and it's easy to understand why it would work (taking advantage of statistically oversold condition for entry into well-regarded stocks).

When you run your backtests are you using the exit condition "close is above upper Bollinger Band(16, 2.5) ? It seems like the exit criteria applied on the backtest has a huge impact on the outcome of these tests.

Also, did you compare the effect of different BB settings on the outcome using the backtest?

thanks in advance,
Dean

FYI, I'm backtesting a similar IBD screening approach using RSI2 below 5 and other conditions as an entry point to see how it compares with the BB approach. I'd like to find an approach that reduces the number of possible trades while maintaining a high win rate and ROI value. Will post the results if I come up with anything interesting or worth sharing.

trendscanner
265 posts
msg #78519
Ignore trendscanner
9/2/2009 9:27:29 PM

One thing that improved backtesting of the IBD BB scan was to eliminate stocks making a large number of new highs before the pullback. The following lines improved the backtesting statistics considerably, at least over the 4/09 through 9/09 time frame. Degree of effectiveness probably varies with specific backtesting time frame.

set {nuhi, COUNT(high reached a new 1 year high, 60) }
add column nuhi
nuhi below 10

hawkeyes002
49 posts
msg #78523
Ignore hawkeyes002
9/2/2009 11:36:48 PM

i dont understand the equity aspect of backtrading? any help

Kevin_in_GA
4,557 posts
msg #78524
Ignore Kevin_in_GA
modified
9/2/2009 11:41:24 PM

The equity summary is simply the total amount of cash you have at the end of the backtest period. You start with $100,000 - a filter is designed to make you money, so it better have a decent equity summary.

The example earlier where all of the trades were profitable but the equity gained was only 1% at the end. So what? Put profits before ego. THAT is why the equity summary should be included in any posting of backtest results - it's what matters most.

billrider321
70 posts
msg #78528
Ignore billrider321
9/3/2009 1:06:44 AM

Nice work Kevin, i was a big fan of testing stuff on ibd100..here is the catch

1. what i find is if you back test on todays ibd100 you are not considering the fact that it has made to ibd100 becasue if has performed well in past.

try this -

take ibd100 list from 12/1/2008 and try your filter between 12/1/2008 to 2/28/2009

i am just trying to point out my research and thought process..would like to get others view.

richo
74 posts
msg #78533
Ignore richo
9/3/2009 6:54:57 AM

Kevin,

I have 3 from the 3 (IBDlists) for today. Is this correct?

CACC
CA
NCIT


Regards


Rich

Kevin_in_GA
4,557 posts
msg #78534
Ignore Kevin_in_GA
9/3/2009 7:44:24 AM

Nice work Kevin, i was a big fan of testing stuff on ibd100..here is the catch

1. what i find is if you back test on todays ibd100 you are not considering the fact that it has made to ibd100 becasue if has performed well in past.

try this -

take ibd100 list from 12/1/2008 and try your filter between 12/1/2008 to 2/28/2009

i am just trying to point out my research and thought process..would like to get others view.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

This was optimized against the IBD top 100 list from 4/6/09

I'm not sure it is fair to test from 12/1/08 to 2/28/09 - that is a very challenging stress test. Also no reason to limit the stock choices to only the top 100 when you can use more stocks that are also highly rated (the consolidated lists I posted have 458 stocks in them, when duplicates between lists are removed).

I am in the process of re-optimizing this against the 458 stock list and using 1 yr ago as the start date - the optimization takes commissions into account, so that the real profitability of a given system is revealed.

StockFetcher Forums · Filter Exchange · Interesting filter - perhaps the best I've seen so far<< 1 ... 3 4 5 6 7 ... 8 >>Post Follow-up

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2016 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus