StockFetcher Forums · Filter Exchange · STOCHASTIC CROSSOVER SYSTEM FOR SDS AND SSO<< 1 2 3 4 5 ... 22 >>Post Follow-up
Kevin_in_GA
4,548 posts
msg #107364
Ignore Kevin_in_GA
8/5/2012 3:46:22 PM

I thought it might be time to start looking at additional market filters based on simple crossover/crossunder behaviors. This filter was the result of several iterations on Stratasearch, starting with the ^SPX, then using leveraged ETFs as trading stocks, and finally to include a trailing stop to provide some form of hazard insurance.

Fetcher[

/*A SIMPLE STOCHASTIC CROSSOVER APPROACH TO MARKET TIMING WTH SSO AND SDS.

WHEN A SIGNAL IS TRIGGERED, ENTER THAT STOCK AT THE NEXT DAY'S OPEN. I DID NOT TEST ENTERING AT THE CLOSE OF THE SIGNAL DAY, BUT THAT PROBABLY WORKS JUST AS WELL.

EXIT IS WHEN THE STOCHASTICS %D(5,1,5) CROSSES BACK ABOVE THE STOCHASTICS %D(5,1,10)

NOTE THAT A 3% TRAILING STOP IS PUT IN PLACE ON ALL TRADES - NO EXCEPTIONS.*/

SYMLIST(SDS,SSO)
SET{GO_LONG, COUNT(STOCHASTICS %D(5,1,5) BELOW STOCHASTICS %D(5,1,10),1)}
SET{GO_SHORT, COUNT(STOCHASTICS %D(5,1,5) ABOVE STOCHASTICS %D(5,1,10),1)}
SET{NULL,0}

ADD COLUMN SEPARATOR
ADD COLUMN GO_LONG {GO LONG}
ADD COLUMN GO_SHORT {GO SHORT}
ADD COLUMN SEPARATOR
ADD COLUMN STOCHASTICS %D(5,1,5) {%D(5,1,5)}
ADD COLUMN STOCHASTICS %D(5,1,10) {%D(5,1,10)}

DRAW GO_SHORT ON PLOT NULL
DRAW GO_LONG ON PLOT NULL
DRAW STOCHASTICS %D(5,1,5) ON PLOT STOCHASTICS %D(5,1,10)

]



Here is how this strategy traded from 1/2/2007 until ths past Friday:


Photobucket


and the stats:


Photobucket


Of note for me is the high monthly Sharpe ratio of 0.4065 (annualized at 1.40). This is unusual given the use of leveraged ETFs - what most people fail to realize is that leverage usually results in a lower Sharpe ratio due to the increased volatility even after the larger gain. Also remember that this included all of the 2008/2009 meltdown.

Happy to share this system with folks here.

Kevin

Kevin_in_GA
4,548 posts
msg #107365
Ignore Kevin_in_GA
8/5/2012 6:59:09 PM

I wanted to confirm these results with the SF backtest data, but for some reason only the SSO trades are showing up. No idea why since this is designed to alternate between SDS and SSO based on their mirror image behavior (the cross below entry signal on SSO corresponds to the cross above exit for SDS, and vice versa).

I'll start tracking this here in this thread at the next crossover.

novacane32000
273 posts
msg #107366
Ignore novacane32000
8/5/2012 8:28:28 PM

Kevin ,Thanks.
Always exciting when you come up with something new.
Have not had a chance to digest this one yet but I like that you incorporated a stop.

tombrown1
61 posts
msg #107369
Ignore tombrown1
8/5/2012 10:38:51 PM

Kevin,
Sorry, but I'm not quite understanding this - pardon my ignorance.

When STOCHASTICS %D(5,1,5) BELOW STOCHASTICS %D(5,1,10),1
- you go long on either SSO or SDS

But then you also short them if it is above? So you are always short one and long the other at the same time?
I guess I'm confused by your usage of the word "short". Are you literally meaning to short either of these stocks, or are you meaning "short" to indicate buying into sds?

I tried pasting your code into a backtest, and had horrible results for the last two years with your exit and the trailing stop. However, I don't think I'm backtesting correctly because I don't really understand what you're doing.

Any help will be appreciated, and if I can understand it a little better then maybe I can be of some assistance in backtesting and optimizing.

Best,

Tom

tombrown1
61 posts
msg #107370
Ignore tombrown1
8/5/2012 10:47:46 PM

Using this as an entry:

SYMLIST(SDS,SSO)
STOCHASTICS %D(5,1,5) crosses BELOW STOCHASTICS %D(5,1,10)

and your exit of:
STOCHASTICS %D(5,1,5) CROSSES above STOCHASTICS %D(5,1,10)

or

3% trailing stop

over the last two years with 2 maximum open positions and 2 trades per day

buying and selling at the open


Your equity summary would've lost about $300 total.



Kevin_in_GA
4,548 posts
msg #107374
Ignore Kevin_in_GA
8/6/2012 8:26:37 AM

Using this as an entry:

SYMLIST(SDS,SSO)
STOCHASTICS %D(5,1,5) crosses BELOW STOCHASTICS %D(5,1,10)

and your exit of:
STOCHASTICS %D(5,1,5) CROSSES above STOCHASTICS %D(5,1,10)

or

3% trailing stop

over the last two years with 2 maximum open positions and 2 trades per day

buying and selling at the open


Your equity summary would've lost about $300 total.
+++++++++++++++++++

I got bad results from SF as well. Two things worth checking:

1. I only saw that SSO was traded, never SDS. Check to see if SDS was ever traded, and if so was it alternating with SSO as it is supposed to be?

2. If you remove the 3% trailing stop, you get 22% return over the last 4 months on the SSO trades. Again, not sure if the stochastics here are calculated in the same manner as they are in Stratasearch.

Also, you should only have 1 position open and 1 trade per day.

Kevin_in_GA
4,548 posts
msg #107375
Ignore Kevin_in_GA
8/6/2012 8:28:11 AM

Also, to make it simpler just ignore the "go_short" column, as this is simply to confirm that the mirror image ETF is sending the opposite signal at the same time.

Nickster8074
53 posts
msg #107381
Ignore Nickster8074
8/6/2012 10:19:02 AM

I backtested this filter by hand using the opening price the day following the signal and had some mighty impressive results. The only difference is that I used 3% stop loss as opposed to a 3% trailing stop loss. After commissions, I came up with slightly more than a 51% return from 4/4/12 to 8/3/12, a four month period. It seems quite choppy, but the results speak for themselves. Excellent filter once again Kevin!

blumberg
27 posts
msg #107383
Ignore blumberg
8/6/2012 11:45:24 AM

The last four months were extremely choppy, therefore a stochastic crossover system has worked very well. I don't see how it would perform well during a trending market.

Kevin_in_GA
4,548 posts
msg #107384
Ignore Kevin_in_GA
8/6/2012 11:59:05 AM

Just look at 2008 for a down trending market and post March 2009 for an uptrend. The graph I posted shows the net equity generated during those times. Both were solid.

Nice thing about this filter is a lot of people can use it at the same time and not skew the results, since the liquidity on these two ETFs is insanely high.

StockFetcher Forums · Filter Exchange · STOCHASTIC CROSSOVER SYSTEM FOR SDS AND SSO<< 1 2 3 4 5 ... 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