|
PureBytes Links
Trading Reference Links
|
Dingo,
Thats why i think the user should have the capability to have a "user defined" benchmark..If one can create it with ATC,and its in the database how would it be any different that choosing the S&P500???
Please excuse me if i am way off as I am just learning the program and am basically a programming numnut.
Allan
----- Original Message -----
From: dingo <dingo@xxxxxxxxxxxx>
Date: Tuesday, July 11, 2006 6:38 pm
Subject: RE: [amibroker] Comparitive Buy and Hold returns
> The composite idea is a good one but I'm afraid if you do anything
> it will
> become a thing on its own (like everything else)...
>
> You'll get requests like:
>
> "How about letting us give you the percentages to use of each ticker?"
>
> "Why can't we just specify a ticker ( or watchlist or ... ) to use in
> building the composite?"
>
> "Could you make the composite xxxx weighted?"
>
> ad naseum.
>
> But then I'm the grumpy one. 8-)
>
> d
>
>
>
>
>
> _____
>
> From: amibroker@xxxxxxxxxxxxxxx [mailto:amibroker@xxxxxxxxxxxxxxx]
> On Behalf
> Of Tomasz Janeczko
> Sent: Tuesday, July 11, 2006 5:53 PM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: Re: [amibroker] Comparitive Buy and Hold returns
>
>
> Hello,
>
> This was the reason why I did not include it when I wrote portfolio
> backtester, but since then
> I was considering adding this option for comparison purposes.
> One idea (as suggested by Allan) is to use benchmark such as
> SP500, but it
> won't be very useful
> when testing some small watch lists of stocks that are little
> correlatedwith large caps.
> Also for some instruments there is no benchmark (see forex for
> example),that's why I am thinking about another solution:
> creating own "benchmark" by using all symbols under test to create
> equalweighted composite
> and test buy and hold strategy on such artificial composite. This
> would give
> you some idea how
> well given strategy performs compared to passive investment.
>
> Any comments?
>
> Best regards,
> Tomasz Janeczko
> amibroker.com
>
> ----- Original Message -----
> From: dingo <mailto:dingo@xxxxxxxxxxxx>
> To: amibroker@xxxxxxxxxxxxxxx
> Sent: Tuesday, July 11, 2006 4:21 PM
> Subject: RE: [amibroker] Comparitive Buy and Hold returns
>
> How would you go about calc'ng B&H on an entire portfolio where
> you used
> scoring, etc?
>
> d
>
>
> _____
>
> From: amibroker@xxxxxxxxxxxxxxx [mailto:amibroker@xxxxxxxxxxxxxxx]
> On Behalf
> Of allansn@xxxxxxxxxxxxx
> Sent: Tuesday, July 11, 2006 8:44 AM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: Re: [amibroker] Comparitive Buy and Hold returns
>
>
>
>
>
> with that said,I can not find buy and hold returns on the new
> report....
>
>
>
>
>
>
> ----- Original Message -----
>
>
>
> From: Tomasz Janeczko <groups@xxxxxxxxxxxxx>
>
>
>
> Date: Monday, July 10, 2006 5:08 pm
>
>
>
> Subject: Re: [amibroker] Comparitive Buy and Hold returns
>
>
>
>
>
> > Hello,
> >
> > 1. Yes
> > 2. Yes from mathematical standpoint however you would NOT be
> able
> > to enter that many positions with limited capital
> > so these results are theoretical only.
> >
> > 3. OLD backtester is phased out. It is there only for people who
> > had some resutls from versions 4.40 or earlier stored somewhere
> and
> > want to reproduce old backtests for some reason.
> >
> > All new code should use NEW portfolio-level backtester only.
> >
> > Best regards,
> > Tomasz Janeczko
> > amibroker.com
> > ----- Original Message -----
> > From: allansn@xxxxxxxxxxxxx
> > To: amibroker@xxxxxxxxxxxxxxx
> > Sent: Monday, July 10, 2006 9:59 PM
> > Subject: Re: [amibroker] Comparitive Buy and Hold returns
> >
> >
> >
> > So,if the system is a losing system and has a an average loss
> as
> > opposed to profit,will the result be apx((approx (number of
> > symbols) * avg. loss ) ??
> >
> > Are the buy and hold stats accurate?
> >
> > Where can i find the definitions for the old backtest??
> >
> >
> >
> >
> >
> >
> > ----- Original Message -----
> >
> >
> > From: Tomasz Janeczko <groups@xxxxxxxxxxxxx>
> >
> > Date: Monday, July 10, 2006 2:59 pm
> >
> > Subject: Re: [amibroker] Comparitive Buy and Hold returns
> >
> >
> > > Hello,
> > >
> > > Please note that OLD backtester is not portfolio backtester.
> > > It backtests each security *separately* and sums up the
> > results
> > > so your profits will be much higher (approx (number of
> > symbols) *
> > > avg. profit )
> > >
> > > Best regards,
> > > Tomasz Janeczko
> > > amibroker.com
> > > ----- Original Message -----
> > > From: "matrix10014" <allansn@xxxxxxxxxxxxx>
> > > To: <amibroker@xxxxxxxxxxxxxxx>
> > > Sent: Monday, July 10, 2006 6:35 PM
> > > Subject: [amibroker] Comparitive Buy and Hold returns
> > >
> > >
> > > > Can someone please tell me where in reports you can easily
> > see
> > > how
> > > > your system did relative to a buy and hold strategy??
> > > >
> > > > I did see some Buy and Hold stats in the Old backtester,but
> > some
> > > of
> > > > the numbers were of by a factor of 10-100..
> > > >
> > > >
> > > > Total net profit: 532108.89 Total commissions paid: 0.00
> > > > Return on account: 6.82 % Open position gain/loss
> > 11931.24
> > > > Buy&Hold profit: 3657774.41 Bars (avg. days) in test:
> > 13351
> > > (250)
> > > > Buy&Hold % return: 46.89% System to Buy&Hold index: -
> > 85.45%
> > > >
> > > > Annual system % return: 10.11% Annual B&H % return:
> 75.32%
> > > >
> > > >
> > > > those numbers posted were on a $100,000 account!!I have no
> > idea
> > > where
> > > > the total net profit number of 532,108.89 came from..And I
> > > really dont
> > > > know how a 532,108 profit only generates a 6.82% return on
> a
> > > $100000
> > > > account
> > > >
> > > > Thanks
> > > >
> > > > Allan
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > Please note that this group is for discussion between users
> > only.
> > > >
> > > > To get support from AmiBroker please send an e-mail
> directly
> > to
> > > > SUPPORT {at} amibroker.com
> > > >
> > > > For other support material please check also:
> > > > http://www.amibroker.com/support.html
> > > >
> > > >
> > > > Yahoo! Groups Links
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > >
> >
>
>
>
>
>
>
|