|
Hi Colin
If the sentence you've picked up on was the only
time I'd ever been wrong I'd be pretty happy. It seems that the wording I've
used is not strictly accurate, and neither does it convey the message I
intended. The essential point to take from the discussion is that an N/A
plot from a latch expression can extend much further than one
would expect. Encapsulating all possible permutations of latch input
signals in one sentence is a difficult task, and my attempt to do
so seems to have come up short.
In my own defence, I must say that there are
several revisions of the document discussing latches, and I can't find the same
statement in the latest version. It seems that you have an excellent grasp of
the subject matter, so you could probably put this form of latch behind you and
use the Equis Forum dll latch function in the future.
ExtFml( "Forum.Latch",N,X,0,0);
This dll doesn't require an "Init" variable, and it
appears to be impervious by the nuances that affect BarsSince(), ValueWhen() and
other similar MetaStock functions.
Regards
Roy
----- Original Message -----
Sent: Tuesday, November 01, 2005 6:21 PM
Subject: [EquisMetaStock Group] Question re invalid period in
latches
I have been struggling to understand the following section of
Roy
Larsen's excellent paper on latches:
He says:
One problem to
be addressed with this simple code is that the output
line (using BarsSince)
can have an extended invalid signal period. In
other words it will return a
value of N/A instead of a legitimate
value. This N/A period will extend from
the first bar on the chart
until the `set' signal is true and the `reset'
signal is valid.
My difficulty is with the last sentence.
As I
understand it, BarsSince(foo) becomes valid the first time foo
is true and
therefore the expression BarsSince(N) < BarsSince(X) will
become valid
when all its constituents are valid i.e. when N and X
are both true for the
first time. If N happens before X, I can
envisage a situation where N is
true and X is valid but X has not yet
been true. In this case the N/A period
will extend beyond the
situaton where "the `set' signal is true and the
`reset' signal is
valid".
Can someone explain to me why the last
sentence should not have read
something like:
"This N/A period will
extend from the first bar on the chart until
the `set' signal and the
`reset' signal have both been true at least
once".
Colin
YAHOO! GROUPS LINKS
|