|
PureBytes Links
Trading Reference Links
|
One thing I've found in working with TS Indicators, Functions is that you
can't trust the data in the Data Window -- it may or may not be the same as
that generated by the calculations in a System. I would assign the results
from your MACD calcs. to Variables (Value1, Value2, etc.) and print them out
to the Print Log, or an Ascii file and verify that they actually contain the
values that you assume they do.
In one case, I found that the Max-Bars setting in a Indicator was
"Auto-Detect" an the same setting in a System was 50 -- this made a major
difference in how the System operated, but you had to print the interim
calculations and compare them with the Study values to discover this.
Despite a lot of mickey-mousing around, it was never possible to get the two
to match.
Hope this helps -- good luck!
---------------------------------------
At 12:27 pm 12/17/97 -0600, you wrote:
>I've seen other posting recently describing problems similar to mine. I
>have a system defined that has as one of the criteria that the
>difference between the MACD and the MACD moving average must be less
>than or equal to zero. Using the formulas from the help file, the
>condition is coded as follows:
> (MACD(close,5,30)-XAverage((MACD(Close,5,30)),30))<=0
>
>When applied to the 3/98 Corn futures chart, an erroneous sell signal is
>generated on 12/23/96 and 12/30/96. The data in the Data Window shows
>the following values for MADiff:
> 12/23/96 0.14
> 12/30/96 0.45
>
>There are no negative values for MADiff within two days on either side
>of these dates.
>I'm using the built-in MACD indicator, but I also entered the above code
>and it gives the
>same values as the build in indicator.
>
>I'm at a loss to explain why this is happening. Any ideas?
>
>Regards,
>Glenn Crumpley
>
>
>
|