Short S&P system?

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
nzbryant
Senior Member
Senior Member
Posts: 29
Joined: Mon Dec 13, 2004 10:43 pm

Short S&P system?

Post by nzbryant » Thu Oct 29, 2009 8:37 am

Hi all. I've been testing various ideas for a short S&P system. The purpose is, if you have a portfolio of stocks, you may want to hedge, at peaks, against market declines like this week.

Ive tried CCI, RSI, ADX, DI+/- and ROC. Nothing really works. Perhaps the answer lies in market internals like advance/decline etc. Has anyone come across a good short system for catching peaks (not a day trading system), or any ideas?

The idea is to be short for a day or up to two weeks at equity market peaks.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Thu Oct 29, 2009 9:55 am

You might try the systems in (book 1) or (book 2). If you code them in Blox, you can use the "Trade Direction Portfolio Manager" which makes it simple to test them as LongOnly, ShortOnly, or LongAndShort systems. Maybe you'll discover that you like their behavior on Long trades too; who knows?

nzbryant
Senior Member
Senior Member
Posts: 29
Joined: Mon Dec 13, 2004 10:43 pm

Post by nzbryant » Thu Oct 29, 2009 11:09 am

Thats a good answer Sluggo - thanks. Landry and Connors swing trading books.
(To be honest I dont have Tblox (I use C#, but will use Tblox when the cost is a smaller percentage of my account)).

danZman
Roundtable Knight
Roundtable Knight
Posts: 117
Joined: Thu Mar 10, 2005 10:07 am
Location: Newport Beach,CA

Post by danZman » Thu Oct 29, 2009 1:54 pm

I've found that a top down approach works well with the S&P 500.

First, determine the intermdiate-term trend. No small feat there.

Then use a counter-trend strategy to sell short on rallies. That's
the easy part. Getting out of your system's way and not screwing
up...a lot harder.

D

cryder
Roundtable Knight
Roundtable Knight
Posts: 101
Joined: Tue Jul 22, 2008 5:02 am
Location: Melbourne, Australia

Post by cryder » Fri Oct 30, 2009 3:55 pm

Have a look at the Pivot Detector Oscillator described in TASC July 09. It codes up pretty easily in TB and may give you a viable starting point.

Rgds

Craig

nzbryant
Senior Member
Senior Member
Posts: 29
Joined: Mon Dec 13, 2004 10:43 pm

Post by nzbryant » Fri Oct 30, 2009 8:28 pm

danZman wrote:I've found that a top down approach works well with the S&P 500.

First, determine the intermdiate-term trend. No small feat there.

Then use a counter-trend strategy to sell short on rallies. That's
the easy part. Getting out of your system's way and not screwing
up...a lot harder.

D
DanzMan, so lets say the intermediate trend for S&P500 is up, as in Sep and Dec this year (09). Then you say use a counter-trend strategy to short on rallies. Yes, well thats exactly what I have been testing for. I couldnt find anything that works.

Getting out of my way is no issue - I execute to the letter - very anal that way.

The issue is that picking tops is almost impossible mathematically. Strategies that do it also short on strong trends that keep on going up!

Sluggo, I checked those books you suggested, thanks. They generally go for pullbacks in trends (so therefore "with the trend"), so they didnt help much. But thanks anyway. Maybe that is the answer, even Connors and Landry cant pick tops to short.

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Sat Oct 31, 2009 5:50 am

nzbryant wrote:The issue is that picking tops is almost impossible mathematically. Strategies that do it also short on strong trends that keep on going up!
I beg to suggest that this is not the issue. Picking temporary tops isn't that impossible. And tops don't have to last forever, just long enough to make money on your average trade. It also doesn't matter if trends keep going up or not - that's what stops are for after all.

The issue is to find a way of picking reversals along with appropriate stops and profit capturing rules that is profitable. Its likely to only work on things that revert a lot - or best anyway. Its also likely to have rules to avoid repeated errors on a trend day.

There are a lot of people who very successfully pick (temporary) tops and take a buck from it. Then they look for the next one which might be higher or lower than the last one ... etc.

davidjohnhall
Contributor
Contributor
Posts: 2
Joined: Sat Feb 07, 2009 2:58 pm

Re: Short S&P system?

Post by davidjohnhall » Mon Nov 02, 2009 4:18 am

nzbryant wrote:Hi all. I've been testing various ideas for a short S&P system. The purpose is, if you have a portfolio of stocks, you may want to hedge, at peaks, against market declines like this week.

Ive tried CCI, RSI, ADX, DI+/- and ROC. Nothing really works. Perhaps the answer lies in market internals like advance/decline etc. Has anyone come across a good short system for catching peaks (not a day trading system), or any ideas?

The idea is to be short for a day or up to two weeks at equity market peaks.
You cannot know where the peak was until it has passed. In effect, traders trying to find peaks only helps the trend traders cause that much more.

Instead of trying to trade or find peaks, jump to smaller time frames and trend trade those time frames. For 2-5 day trades look at 10 - 15 bar Donchian channels on 15, 30 and 60 minute charts.

A usual trend on those time frames lasts anywhere from 2-10 days. I have found it useful for hedging if that's what you're looking for.

nzbryant
Senior Member
Senior Member
Posts: 29
Joined: Mon Dec 13, 2004 10:43 pm

Re: Short S&P system?

Post by nzbryant » Mon Nov 02, 2009 5:56 am

davidjohnhall wrote: You cannot know where the peak was until it has passed. In effect, traders trying to find peaks only helps the trend traders cause that much more.

Instead of trying to trade or find peaks, jump to smaller time frames and trend trade those time frames. For 2-5 day trades look at 10 - 15 bar Donchian channels on 15, 30 and 60 minute charts.

A usual trend on those time frames lasts anywhere from 2-10 days. I have found it useful for hedging if that's what you're looking for.
Davidjohnhall

You are right. Trend trading on shorter time frames would help pick tops on a daily timeframe. I cant do that as have a day job (dont want to robo-trade) but you are right.

nzbryant
Senior Member
Senior Member
Posts: 29
Joined: Mon Dec 13, 2004 10:43 pm

Post by nzbryant » Mon Nov 02, 2009 5:58 am

Kiwi wrote:
nzbryant wrote:The issue is that picking tops is almost impossible mathematically. Strategies that do it also short on strong trends that keep on going up!
I beg to suggest that this is not the issue. Picking temporary tops isn't that impossible. And tops don't have to last forever, just long enough to make money on your average trade. It also doesn't matter if trends keep going up or not - that's what stops are for after all.

The issue is to find a way of picking reversals along with appropriate stops and profit capturing rules that is profitable. Its likely to only work on things that revert a lot - or best anyway. Its also likely to have rules to avoid repeated errors on a trend day.

There are a lot of people who very successfully pick (temporary) tops and take a buck from it. Then they look for the next one which might be higher or lower than the last one ... etc.
Thanks Kiwi. I tried optimising stops and targets, but still had only a low single digit CAGR and up to 15% drawdown. There were times (05 I think it was) when the S&P500 hit high RSI, CCI, ROC and just kept on going. Maybe I need rules, as you suggest, where a trade is taken or not based on the success of the last trade.

Post Reply