[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Walk forward optimization



PureBytes Links

Trading Reference Links

Jim and Alex:

Walk forward optimizing and the use of a spread relationship for a system
screen are totally different topics which should be considered separately,
imho.

In a short period of a few months of testing walk forward optimizing of a
successful system versus a set of parameters that were not changed, I found
that the walk forward optimizing did much worse than the 2 yr optimization I
had done and then left unchanged.  This is not enough length of the WFO to
be certain, but the info was enough for me, at the time, to consider other
projects as more important.

As far as the topic of the use of a spread relationship as a system screen,
it certainly may be and can be helpful at times when the seasonality is in
favor of the system parameters.  However, when your system parameters are
contraseasonal, you will have bad results.  Doing testing of the seasonal
relationships of the spread over a long period (Moore Research uses 15 yrs)
could be programmed into a system and would be much more useful as a system
screen than one without seasonality considered.

Thanks,

Don

----- Original Message -----
From: "Jim Bronke" <jvbronke@xxxxxxxx>
To: "Alex Vinogradov" <avinojr@xxxxxxxxx>; <omega-list@xxxxxxxxxx>
Sent: Monday, January 14, 2002 11:50 AM
Subject: Re: Walk forward optimization


For example, in his book Cybernetic Trading, Ruggerio mentions a
relationship between Crude oil and the dollar. The solution is to add the
dollar as a data2 if you see that you can add that to your system in a way
that increases your return.

Jim Bronke
Phoenix, AZ



----- Original Message -----
From: "Alex Vinogradov" <avinojr@xxxxxxxxx>
To: <omega-list@xxxxxxxxxx>
Sent: Monday, January 14, 2002 10:40 AM
Subject: Re: Walk forward optimization


: Hi Jim,
:
: Could you elaborate on that a little bit?  I've just
: spent a bit (more that I should have) of time working
: with a walk-forward optimization protocol, only to
: conclude that it produces little more than confusion.
: What are your thoughts?
:
: Thanks in advance,
:
: Alex Vinogradov
: Miami, FL
:
: --- Jim Bronke <jvbronke@xxxxxxxx> wrote:
: > It is better to include new relationships than to
: > walk forward old ones:
: >
: > Jim Bronke
: > Phoenix, AZ
: >
: >
: >
: > ----- Original Message -----
: > From: "David Folster" <mr_bond@xxxxxxxxx>
: > To: <omega-list@xxxxxxxxxx>
: > Sent: Sunday, January 13, 2002 9:06 PM
: > Subject: Walk forward optimization
: >
: >
: > : Anyone have an opinion, good or bad, on walk
: > forward optimization?
: > :
: > : Thanks,
: > :
: > : David
: > :
: > :
: >
: >
:
: