|
PureBytes Links
Trading Reference Links
|
A response regarding "voting" or "concensus" based systems:
- no doubt code would be more complex......but it's not that difficult to
assign 2 TrueFalse variable arrays (or numericseries arrays with 0/1 values)
that would contain a true or false (1/0) for EACH indicator to be applied in
the system.....one for the long side, the other for the short side.
- the system would decide on what a "majority" really is defined as.....40%,
60%, 80%, etc.
- the system would go long or short or exit long or exit short based on the
majority "rules".
My biggest concern would be the number of trades that would evolve because
of the potential for conflicting signals.......
you may end up with 1 trade per ten thousand bars or 1 per month ....or year
!!
> -----Original Message-----
> From: Bob Fulks [mailto:bfulks@xxxxxxxxxxxx]
> Sent: Sunday, November 21, 1999 11:48 AM
> To: Larry McBride
> Cc: Omega List
> Subject: Re: Composite Timing Models
>
>
> There was a system like this published in TASC (January 1995) called
> "Insync Index". It used probably ten different common indicators
> (MACD, RSI, etc) and entered the market when some majority "voted" to
> be in. Greg Morris sold the code for this system for a nominal
> amount. I just checked his web site and it still seems to be
> available should you like to play with it.
>
> http://www.murphymorris.com/indicators_systems.html
>
> As I recall, it was fairly robust for a system based upon such simple
> indicators.
>
> The problem with some such systems is that one signal can be short
> while another is long so it is sometime difficult to write the code
> so that the signals do not interact incorrectly.
>
> I have found that it was better to have one main trading signal that
> is based upon some fundamental market characteristic that is
> repeatable. Then, you can often improve the performance of that
> system by using other signals to filter out trades to improve $ per
> trade, % profitable, etc.
>
> Bob Fulks
>
>
> At 6:09 PM -0800 11/20/99, Larry McBride wrote:
>
> >Here's a question about the development of market timing models I haven't
> >seen discussed clearly. It has to do with the development of
> robust models
> >from a statistical perspective.
> >
> >Assume you have three different timing signals you've developed
> separately
> >that are based upon reasonably different market parameters or
> else different
> >ways of modeling the market. For example, you may be trying to
> time the otc
> >composite for mutual fund trading and you develop three separate timing
> >signals:
> >1) an percent swing system based upon and down price action.
> >2) an indicator based system that totals a stochastic and rsi oscillator,
> >then trades off the combined oscillator.
> >3) a relative strength signal that trades based upon the
> relative strength
> >of the otc versus some other index.
> >
> >Now, you develop each of the three signals separately based upon
> in-sample
> >data, accepting them if they look reasonable on out of sample data. Of
> >course some look better than the others in out of sample
> observations. The
> >ones that look best out of sample are very often not the ones that looked
> >best in sample.
> >
> >Now, you build a composite timing model that does something like buy when
> >any 2 of these signals are long and sell when less than 2 of
> them are long.
> >You determine the most effective voting process by looking at
> the in-sample
> >data only. Almost invariably, it has looked to me like you would
> have been
> >better off trading such a composite in the out of sample data
> than you would
> >have been in trying the pick the best individual signal from in
> sample test
> >results. Although this is a heuristic observation, I've seen it
> enough to
> >make me think there is something to it. It has remained
> apparently true as
> >well if I use three different time periods instead - in sample, out of
> >sample & independent verification.
> >
> >My question is the following, which is really statistical in nature. Is
> >such an approach (composites of separate timing signals) likely
> to be more
> >robust in the future since the composite model is less dependant upon the
> >success of any given signal, or is it likely to be less robust since more
> >parameters (the total of those involved in all signals plus the number of
> >signal defined in the voting process) are nonetheless involved in the
> >composite model?
> >
> >Thoughts anyone? Thanks for your ideas.
> >Larry
>
|