|
PureBytes Links
Trading Reference Links
|
I would like to see a product that can backtest complex trading strategies
on multiple securities.
An example of a simple, basic system idea to backtest (I would like to be
able to test and trade much more complex strategies, but even this is
breaking new ground):
Stock universe: Nasdaq stocks with float > 10 million and 50 day adaptive
moving average trading volume > .5 million.
If a technical indicator for bull/bear/flat in the appropriate timeframe for
Nasdaq = "bull", then
On daily bars, find the 10 stocks whose highs are higher than those of any
day in previous 200.
Buy N dollars of stock if/when that high is penetrated intraday within some
number of days. N dollars is dynamically determined based on number of
realizable trades in recent time period.
Go flat each stock when momentum dips below optimized value or when it
reaches a logical technically determined stop level or when a juicier trade
comes along.
End
Opposite for bear market.
I would like all the numbers in this trading stategy to be adaptive and
dynamically determined, and I would like FAST optimization capabilities, as
well as AUTOMATIC data segmentation and out of sample testing capabilities.
Packages like Biocomp Profit and SirTrade have some of those capabilities.
I would like to be able to use options to hedge or to trade. I would like
to be able to use complex indicators. I would like to be able to include
fundamental data such as earnings, float, Fed interest rate changes. I
would like to be able to integrate AI technologies (neural nets, fuzzy
logic, etc).
The Tradestation concept is primarily lateral and one-dimensional: the
movement of one security through time. I would like to expand this to the
vertical dimension of a large universe of security types and securities
(including sectors and sub-sectors of stocks), and to the depth dimension of
all available quantifiable data (not just price and volume).
I assume platforms exist that can do this, perhaps proprietary ones, exist
for institutions with lots of capital. If they exist for individual
traders, I haven't heard of them.
By the way, Mark, what is the $70K/year platform, and does it do any of
this?
-----Original Message-----
From: Mark Brown [mailto:markbrown@xxxxxxxxxxxxx]
Sent: Saturday, October 06, 2001 3:56 PM
To: OMEGA-LIST
Subject: Re: [JurikMA] BEST TRADING SOFTWARE
Hello Patrick,
i will agree, and probably this new product will fail commercially
like traderware did. though it was mostly brought down from internal
let's just say "weirdness". however this product may not need any
customers other than the few who use it. it does not have anyone in
control of it that does not make a living trading, therefore it has
succeeded in it's own rights. there will be no compromises made in
this new product to satisfy the marketing department and take pressure
off the programming dept.
ps i would like to see some "set it apart from ts ideas" thrown
around. make it hard, make it impossible, what the hell. m
PW> I would argue that from a marketing perspective that one reason
TraderWare
PW> failed to penetrate the trading software market is that it failed to
PW> successfully differentiate itself from TradeStation and never created
the
PW> burning need for consumers to buy.
PW> Patrick White
|