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

Fw: Easy Language Math Precision Wrong?



PureBytes Links

Trading Reference Links


 ----- Original Message -----
 From: The Omega Man <editorial@xxxxxxxxxxxxx>
 To: Mark Brown <markbrown@xxxxxxxxxxxxx>; Omega List
<omega-list@xxxxxxxxxx>
 Sent: Saturday, June 19, 1999 6:25 PM
Subject: Re: Easy Language Math Precision Wrong?

> I've said it before and I'll say it again:  This debate over precision is
> ridiculous.
>
 > First, the input data is imprecise.  Prices are set by a bunch of
> screaming
> banshees in a "pit".  Do we need to go out to the nth decimal place to
> decide what the price is?  Or is a good approximation (which is all that
> the
> price spotters can give us anyway) good enough?  I say that it makes no
> sense to run calculations out to the nth decimal place when the input data
> is +/- 0.1 points.

 Prices (H, L, O, C), Volume, UpTicks and DownTicks in TS 4 and TS 2000
appear to be stored as LONG which provides 10 significant digits.   Omega's
Server was designed to handle the needed precision for example in the
occasional Currency traded in 1/100,000.   Volume appears to be able to
store up to 2.1 billion shares per day.

> Second, extra precision is not needed for trading.  What's needed for
> trading is not more precision, but less.  We need "fuzzy" approaches, not
> precise ones.  Carrying calculations out to the nth decimal place in a
> fuzzy
> environment is ridiculous.

You are correct in one sense - it's not a matter of carrying calculations
out to a large number of decimal points.   The problem is that some INTERNAL
calculations - such as multiplication or raising prices to a power require
more than 6 significant digits.

In TS 4 Omega's  "solution" when there are "overflow" exeptions caused by
working with a maximum of 6 available digits is usually to crash
TradeStation Charting, rather than handle the exception properly.
Sometimes the TS 4 Server has to be re-booted to completedly clear the crash
of TradeStation Charting.   Obviously this isn't the type of issue Omega
seems to care about.

I have not tested the "overflow" problem in TS 2000 since I haven't been
able to convert my TS 4 data correctly.   (No - Omega doesn't seem to care
about that either.)