|
PureBytes Links
Trading Reference Links
|
First, I'd like to thank all of you who responded to my problem which led to
this thread. I've included my original message at the end of this post. It
turns out that before my initial post I had tried or looked into all of the
potential 'fixes' which people later suggested .... with the exception of
Jimmy's suggestion to rebuild the workspace and charts. Unfortunately,
Jimmy's suggestion didn't seem to work either. New charts showed the same problem.
Today, however, I stumbled across the answer to the problem and I feel a
bit 'stupid' that I hadn't seen it before! It turns out that for some reason
the chart for which I had the initial problem was set up as semi-long scaling
rather than linear. And when I set up the new charts, they were also
getting set up as semi-log.. Once I changed that, it looks like things are working
the way I think they are supposed to. And it appears like I now have things
set up (again!) so that new charts will be set up as linear. I have NO idea
what I had done to have the particular charts set up as semi-log because I
don't recall ever using semi-log scaling. That, unfortunately, shall remain
an unsolved mystery. I am just pleased that this little problem has joined
the list of the 'solved'!! Thanks again to all who tried to help! :)
Lee Scharpen
================================================================
In a message dated 10/19/2005 9:13:22 AM Pacific Daylight Time,
_LScharpen@xxxxxxxx (mailto:LScharpen@xxxxxxx) writes:
I've been using TS products for a number of years and TS2000i for the last
5. A few days ago I noticed some 'strange' behavior in the charting of a
Simple Moving Average. It was significant enough that I thought I'd check
to see
if anybody else may have seen this or might see it in the future. Let me
explain.
Let me first give the specifics of the chart. I don't think these
'specifics' matter much but they at least provide a concrete starting
point. My chart
was a 1 min. bar of ES Z5. On this I had Bollinger Bands (20, 2 stdev)
and
a 20 period SMA. I then happened to notice that it appeared that the 20
period SMA was NOT in the middle of the upper and lower Bollinger Bands.
This
was confirmed in several ways. The 'visual' method was done by putting
ANOTHER 20 sma indicator on the chart and observing that the two different
20 sma
did NOT overlay each other. However, if I then turned the sma status OFF
and then back ON, the discrepancy 'went away'. Over the course of time,
the
discrepancy would then slowly drifrt back into the chart.
It's a real pain in the ass to not be able to trust something simple like a
moving average. I've given enough detail so that it would be relatively
simple for somebody else to try to reproduce what I've seen. I'd sure
appreciate
hearing form anybody who tries the experiment and, more important, from
anybody who has seen this problem and understands how to 'make it go away'
|