+/0/- : Signs of Differences In EIS Stats

[Leroy Quet]

Just curious.

If we form, based on any sequence {a(k)} in the
EIS,  a sequence of +s, 0s, and -s where every
sign is the sign of (a(k)-a(k-1)), then which
patterns of signs are the most common?

I am absolutely certain that the sequence of all
+s is the most common. But what is the second
most common sign-sequence?
The third most common, etc?

There should be some ground-rules for this
search.
Since sequences have differing numbers of terms
maybe we should study only the first, say, fifty
or 100 terms (of those sequences with at least
that many terms.)
Also, as far as offsets are concerned, maybe we
should study only how sequences are listed in the
database (with the sign of the difference between
the first two listed terms being the first term
of the sign-sequence for that sequence, whatever
the indexes values).

Also, if we study fewer/more terms of all
sequences having that number of terms, how will
this affect the stats?
(I have a feeling that the number of terms
studied in each sequence greatly affects the
statistics.)

Hopefully this would be easy for someone to check
with their computer.

Thanks,
Leroy Quet





      ____________________________________________________________________________________
Never miss a thing.  Make Yahoo your home page. 
http://www.yahoo.com/r/hs