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

Re: Back testing values



PureBytes Links

Trading Reference Links


----- Original Message -----
From: Al Taglavore <altag@xxxxxxxxxx>
To: <metastock@xxxxxxxxxxxxx>
Sent: Tuesday, April 20, 1999 2:48 PM
Subject: Re: Back testing values


| Rick,
|
| ".......That said, your EOD data provider really should default to
| reporting the
| > close from the stock's primary exchange
|
| ===I checked with BigCharts, Yahoo, Bloomberg and they all reported the
| "composite price".  Only my broker Brown & Co. showed the correct price.


Now you've got me courious Al. I'll have to check into this with my
individual stocks further. I trade mainly the SPY and DIA so I don't have to
worry much so I guess that's another solution!

| But my original question is still unanswered.  How much of this EOD data
is
| different from the actual trading price and how do different traders
handle
| the testing of systems and development of systems when the data has such
| descrepencies.

Thought I did Al with my SPY system example. But if your trading system
signals are indicator dependent and the data tests in the Downloader don't
do the trick for you  but you still want the verification automated, then
you'll need to build a testing mechanism in Excel. It has a full set of
features to handle this. If you are trading tick data, my advice is not use
a price dependent indicator for your exit unless you can run the your RT
system in Excel and are proficient in Excel programming. Use a time limit
for an exit as an example. For an EOD system, after you've gotten your
entry, any errors that are missed shouldn't be a big trading concern so long
as you have your stop-loss or sell orders in place like any off-floor trader
should. Frees you from staring at quote screen too and you can have more
trades on simultaneously to take advantage of any opportunity factor in your
system.

Hope this was a better explanation.
cheers,
Rick