|
PureBytes Links
Trading Reference Links
|
Hi Eric,
> am facing the problem of how to evaluate the exit efficiency. Should
> I set an arbitrary entry strategy as a baseline or should I use the
> prior indicator of the best entry efficiency as a baseline?
> Do you have any suggestion?
The trick is to keep the indicators' entry and exit strategies
independant, and keep them independant from the volatility exit. If
the baseline is something you find in a magazine somewhere, then you
have such an independant strategy as baseline. You should not use the
best indicator as baseline, but some arbitrary one, as the best
probably contains some optimisation already and you are mixing things
and no longer have an apples and apples situation.
I am not sure how the volatility exit you use works, but understand
this is just to pick a good exit once the actual indicator you want to
test opened a trade, and vice versa. Here you could use, in stead of
the volatility exit model, the max profit system. Then the exits will
be on the best points - turning points, but the entries will not and
you will be testing the efficiency of those entries, and vice versa if
you test the exits.
We developed a piece of software that you can use to specify a
percentage, say 10%, and the max profit will exit at the top if the
subsequent fall is more than 10%. Thus you get the best entry and
exit levels possible, ignoring small (10%) noise. We play around with
this 10% level - we do not optimise it - but look at things like
brokerage, margin requirements and risk appetite and so on to guide us
in picking this level. It should not exceed the loss that you can
easily carry in practise, so if your margin allows you no more than a
1% loss, then this suggests the level. We also eye ball the resulting
signals to make sure that we are picking out tops and bottoms and are
ignoring noise. Once we are happy with this, we use it in subsequent
work.
You can do this easily in Metastock. Say you pick 10%, then you use
the MS max profit system, but be sure to enter this value as the
commission. I am not sure if you have to enter this in each of the
entry and exit, or half of this in each of the entry and exit fields.
If you can construct such entry and exit levels, then you can use them
to test all those indicators. Say you work on the entry efficiency.
Each indicator picks the entry level, but the exit comes from this
max profit system. In the end, you are testing the efficiency of the
indicator RELATIVE to the others. Note that you may get very good
results, as the say exit levels in this case are the best. This is
just a relative test and, unless you can attain such perfect exit
levels, you must ignore the values, but look at the relative positions
of the indicators only. The best one has the best entry levels etc.
It is a bit tricky to specify these optimal levels once you have
constructed them. I think you will have to hard code them, so I hope
you are testing this on only one ticker! Here is a good example, from
jose's website, of hard coded entry and exit levels that may help you
in constructing this in Metastock:
http://metastocktools.com/ADT/ADT05sig.txt
Regards
MG Ferreira
TsaTsa EOD Programmer and trading model builder
http://tsatsaeod.ferra4models.com
http://www.ferra4models.com
--- In equismetastock@xxxxxxxxxxxxxxx, chichungchoi <no_reply@xxxx> wrote:
>
>
> Hi MG Ferreira:
> Thank for your suggestion, in order to compare only the performance
> of entry signals for many indicators, I use a volatility exit
> strategy for all of them, so I can compare only the entry efficiency
> among indicators. After that, the highest entry efficiency can be
> determined based on an arbitrary level of volatility exit strategy.
> The problem I am facing is how to determine the appropriated level of
> volatility exit strategy after that, if I change the level of
> volatility, then the prior entry efficiency for each indicator will
> be changed too. I need to determine both entry and exit efficiencies
> independently, but they both are influenced each other. Therefore, I
> am facing the problem of how to evaluate the exit efficiency. Should
> I set an arbitrary entry strategy as a baseline or should I use the
> prior indicator of the best entry efficiency as a baseline?
> Do you have any suggestion?
> Thank you
> Eric
>
>
> --- In equismetastock@xxxxxxxxxxxxxxx, "MG Ferreira" <quant@xxxx>
> wrote:
> >
> > Just a suggestion, if you are going to test a lot of indicators,
> > use one of the simpler ones as a benchmark or use one that you know
> > about, say that was published somewhere. Let's call it the baseline
> > system. Then you can see if the others outperform or underperform
> > relative to the baseline. Also, use the baseline to determine the
> > stop loss and use this for all the rest.
> >
> > Regards
> > MG Ferreira
> > TsaTsa EOD Programmer and trading model builder
> > http://tsatsaeod.ferra4models.com
> > http://www.ferra4models.com
> >
> >
> >
> >
> > --- In equismetastock@xxxxxxxxxxxxxxx, chichungchoi <no_reply@xxxx>
> wrote:
> > >
> > >
> > > Does anyone know how to select the most appropriated level of
> > > volatility stop loss indicator for a group of indicators?
> > > There are a group of indicators to trigger their buy signals only
> and
> > > their exiting strategies are based on a volatility stop loss
> > > indicator. For each indicator, their stop loss indicator can be
> > > optimized for different levels of volatility, but there will be
> > > difficult for comparing their buy signals' performance, since
> their
> > > exiting strategies are using different levels of volatilities or
> > > comparing apples with oranges. If I set an arbitrary level for
> the
> > > volatility stop loss indicator for all indicators, then their buy
> > > signals' performance can be compared, since we are comparing
> apples
> > > with apples. However, I get no idea how to determine the most
> > > appropriated arbitrary level of volatility for the stop loss
> > > indicator, does anyone have any great idea?
> > > Thank you
> > > Eric
------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/Tcy2bD/SOnJAA/cosFAA/BefplB/TM
--------------------------------------------------------------------~->
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|