StockFetcher Forums · Filter Exchange · STOCHASTIC CROSSOVER SYSTEM FOR SDS AND SSO<< 1 ... 5 6 7 8 9 ... 22 >>Post Follow-up
Kevin_in_GA
4,552 posts
msg #107588
Ignore Kevin_in_GA
8/13/2012 4:32:15 PM

I'll rerun this with no trailing stop from 1/2/2007 until 8/3/2012 and then also look at trailing stops from 1 to 10%. Will post the Stratasearch results later tonight (numbers only, not charts).

The point here is that this system performs well - the debate is on exactly how well.

Using the backest feature on ProdigoRTS (www.prodigiorts.com, a great robotic trading tool), I got the following trades and results. Commissions are calculated at 0.01 per share.


Photobucket


The trades here are all roughly fixed at $49,000 instead of compounding from one trade into the next. So what I did was to use the Net P/L % data and multiplied to get the result. Using this rollover approach (which is what Stratasearch does) the system is up ~61% so far this year.

novacane32000
273 posts
msg #107589
Ignore novacane32000
8/13/2012 4:33:15 PM

tconte.Try only taking SSO trades when the market is flat or trending up and SDS trades when the market is flat or trending down.Use the 50(ma) as an indicator to tell you which way the trend is going.
As an example looking at the SPY chart you would have only taken SSO trades from Oct 2011 through May of 2012.
From May 2012 through end of July 2012 the 50ma is down then flat so you take SDS trades only when the trend is down then both SDS and SSO trades as it flattens out
This month the 50 ma turned up again so you would only take SSO trades.

I actually did this with the 75 (ma) and got improved results from Jan 2011 to present . I suspect the 50ma may yield even better results .

tconte
13 posts
msg #107592
Ignore tconte
8/13/2012 8:03:46 PM

Thanks for your great suggestion, novacane. After testing trend with the SPY MA(50) (Aug2008 to Mar2009 down trend or flat; Apr2009 to Apr2010 up trend or flat; May2010 to Aug2010 down trend or flat; Sep2010 to May 2011 up trend or flat;Jun2011 to Sep2011 down trend or flat; Oct2011 to Apr2012 up trend or flat; May 2012 to Jul2012 down trend or flat ), the results are way better. The 2 year period for getting out of the slump has been shrunk to 8 months, and the end capital has increased to $700 k.
check it out here:
https://docs.google.com/spreadsheet/ccc?key=0Ah-vGIsKIZandFN2ZWpIb1p6bzRUNmU3bzZERW9mTWc#gid=1

jackmack
334 posts
msg #107593
Ignore jackmack
8/13/2012 8:09:01 PM

Kevin
I too am using Prodigio - can I get a screen shot
of the rules for this Stochastic cross system for the RTS?
There are some true limits to any system but
what is great on Prodigio is if you put the rule
to alert you when a condition triggers your rule to buy or sell
it is awesome to see real time.

I have asked the support group at Prodigio (they
are a great group) about rolling ones gains
from one trade to the next to allow stacking one position in a rolling manner
and will post here the reply I get from them.

This looks awesome Kevin - great work as always.
Thank you for sharing.

Eman93
4,659 posts
msg #107594
Ignore Eman93
modified
8/13/2012 8:58:25 PM

johnpaulca
6,121 posts
msg #107512
- Ignore johnpaulca 8/10/2012 12:04:06 AM

you need to catch earlier in the cycle, nothing wrong with the strategy....cycle is long in the tooth and market is mean reverting.
=======================================================================================

Like this?

Image and video hosting by TinyPic

Eman93
4,659 posts
msg #107595
Ignore Eman93
8/13/2012 9:08:17 PM

Its not about when its about where..............

http://stockcharts.com/public/1424685/chartbook

On the left of the chart is all the historic major support and resistance price levels this is all you should need to trade this system.

Pull up a 5 year or 10 year spx and add all horz price lines to your charts and have a look.

long above short below, brain dead simple...... I would use the 20 day instead of the 50 for back test... but would have the 20 be above the 50.


buy the dip and sell the rip...


bjfeen
24 posts
msg #107596
Ignore bjfeen
8/13/2012 9:11:16 PM

tconte, you were correct, there was an error in my spreadsheet -- I was not correctly calculating the 5 day low. I've updated the spreadsheet and the returns improved drastically. I'm still working on including the 3% stop loss...

Kevin, if the 3% stop loss triggers but %D(5,1,5) is still below %D(5,1,10) indicating long, would you buy again at the next day's open or would you wait until the next cycle?

I've updated the shared spreadsheet available in an earlier post with the correct 5 day low. The spreadsheet is currently showing a return of ~800% with this method (without stop loss), but I want to go back and check some of my math.

Fantastic system Kevin and thank you for all your work!

novacane32000
273 posts
msg #107597
Ignore novacane32000
8/13/2012 9:13:41 PM

Wow.Nice results tconte. Thanks for taking the time to verify the results. Lets hope this works moving forward.
And thanks of course to Kevin for this filter.

blumberg
27 posts
msg #107598
Ignore blumberg
modified
8/13/2012 9:36:54 PM

tconte: thx for doing this. Here's my concern: it's easier to spot this "trend" in retrospect, by having the benefit of hindsight.

For example, you write that from Aug. 2008 to March 2009, the trend was "down or flat." Then, starting on April 2009, you say the trend is "up or flat."

However, if you were looking at a chart of the SP500 at that time (April 2009), there was nothing to suggest that the 50DMA was now "up or flat."

In fact, the slope of the 50DMA as of April 1st, 2009 was less "uptrending" than it was 3 months earlier, in Jan. 2009.

Using the 20DMA crossing over a 50DMA as Eman suggested would take away some of the subjectivity of this experiment. However, I wonder if that would have created a lot of whipsawing at times. It would be interesting to backtest it this way if you have the time.

Greatly appreciate yours and everyone's effort here, and ofc thanks to Kevin for being so generous and sharing this.

Kevin_in_GA
4,552 posts
msg #107599
Ignore Kevin_in_GA
8/13/2012 10:06:07 PM

Kevin, if the 3% stop loss triggers but %D(5,1,5) is still below %D(5,1,10) indicating long, would you buy again at the next day's open or would you wait until the next cycle?

++++

No - the entry signal is the cross below, not simply being below. You would wait for the next cross and play that trade.

StockFetcher Forums · Filter Exchange · STOCHASTIC CROSSOVER SYSTEM FOR SDS AND SSO<< 1 ... 5 6 7 8 9 ... 22 >>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