|
PureBytes Links
Trading Reference Links
|
Hey Guys,
you keep me busy. I can not write as fast as you all respond :)
Gary is one who did look at WLD2 and I spend a few hours (or was it days
:)) with him. I respect him as some one who knows TS and is trying to
get to know WLD2.
Gary, I have a few replies for you here:
-----Ursprüngliche Nachricht-----
Von: Gary Fritz [mailto:fritz@xxxxxxxx]
Gesendet: Montag, 18. November 2002 07:04
An: omega-list@xxxxxxxxxx
Betreff: Re: Replacement for 2000i
Leslie Walko wrote:
> I agree with Pierre. Could not have said it better myself.
> I happen to own Wealth Lab.
I wouldn't go quite as far as Pierre did, but I do agree with a
number of his points.
I also own Wealth Lab. I find it useful for testing systems on
baskets of commodities, something that TS provides almost no
support for. I pieced together a hack solution in TS based on
dumping information to files, reading it into Excel, combining
equity curves, etc etc. But that's extremely slow and painful,
and it's very difficult to implement proper money management
techniques.
[VK>>]
That is a good start. Of course it is difficult to change a software
after all this years.
Other than that, I have to say I prefer TS in almost every
respect. Take that with a grain of salt, as I haven't spent very
much time with WL and I'm definitely no expert on it, but a
number of things about WL really bug me. A few examples:
* The interface model is, in my opinion, poorly designed. There
are totally separate interfaces (with similar but different UI's,
different ways of determining defaults, different reporting
outputs) for testing a single market vs. testing multiple markets
with MM (position sizing).
[VK>>]
Well, at least WL has the ability to do test multiple markets. TS does
not have to care about a problem like that.
* Because of this separated design, you can e.g. optimize a
system on a single market, but you CANNOT optimize it on a basket
with MM.
[VK>>]
You can optimize it on a basket using simple money management like the
one that TS is using, correct? You are just not able to optimize 100
markets with a system and a particular MM rule. In any case I think that
this would not give you a real result. As far as I know there are some
plans to include that in a future version, you might even know better?
* WL stores default input values and optimization ranges in the
source, in a rather kludgy comment form. There is NO WAY to
specify a particular value for a specific instance, the way TS
lets you apply a system to two different charts with different
inputs. The only way to simultaneously basket-test a system with
two different sets of inputs is to save two different source
versions.
[VK>>]
I am not sure if I understand correctly. But lets say you optimize a
system on 1000 symbols and you wish to apply the optimum value of each
symbol to each individual stock you just do a right mouse click and pick
the appropriate command. Next time you open up that symbol with this
system it will have the values assigned from the optimization. Did you
talk about that feature?
* WL was originally designed as a stock-trading platform.
Support for futures was tacked on later, and it's not pretty.
The platform has no concept of whether a particular symbol is a
stock, future, index, etc. You have to switch the ENTIRE
PLATFORM from "Stock Mode" to "Futures Mode," which will then
cause it to look up the symbol root in a Futures Symbol Manager
list for the BPV, margin, etc. If you trade only stocks or only
futures then it's not too bad, but it gets a bit messy if you
want to do both.
[VK>>]
When I back test stocks and futures in a basket I just make sure that
the symbols have different names like C and C. If the symbols don’t
match then WLD2 will take symbols that are in the Futures-Symbol-Manager
as futures and anything else as stocks.
* WL has very poor support for keyboard shortcuts. For those of
us with tendonitis in the mouse fingers, using WL for any
significant period of time is quite literally a pain.
[VK>>]
I guess you win here, I never use them so I can not comment.
Not sure whether I like to hear more or whether we should first settle
the issues above? :)
Volker
There's more, but you get the idea.
To WL's credit, it has a thriving user community and Dion
Kurczek, the developer, is extremely responsive to user feedback.
When users find bugs in WL, Dion generally has a fix out within a
week or two. (Wouldn't you love to see THAT from TRAD!!!) I've
discussed most of the above points with Dion and he implied he
will address some of them in the major revision he's working on
now. But there's no way he can respond to every request from all
of his users. He also has the website-based version of WL, and
some of his design choices are limited by a desire to maintain
compatibility between the web version and the desktop
application.
Pierre said:
> Ease of use, stability and reputation is more important in such
> a small market niche.
For once, I agree 100%.
Too bad TRAD doesn't.
TS excels in ease of use in most areas. But the (lack of)
stability of many of their products is near-legendary, and they
have destroyed their reputation with a lot of their former and
potential customers. Pierre, you can insult people and dodge the
issue all you want, but the fact remains that there are a LOT of
people who like TS, and who would love to stick with it, but they
don't trust (or just plain *hate*) TRAD and refuse to do business
with them. That's the main reason people leave TS and go to
platforms like WL, NeoTicker, you name it.
If TRAD listened to their customers, and responded to bug reports
and feature requests the way Dion Kurczek does, they would
utterly obliterate the competition. Their platform is, in most
ways, superior to almost everything else on the market that I'm
aware of. Too bad their customer interaction drives away so many
paying customers.
Gary
|