Mike,

If you're values are more or less 'normally' (in the statistical sense)
grouped around a mean, then you could calculate the standard deviation for
each list and exclude those records which fall without a distance x*SD from
the mean where x is a factor (e.g. 2) which you will have to decide on.

Richard Burkmar.

Mike Atkinson wrote:

I have a series of lists of between 25 and 150 records each  with a
distance field from a known point. (these are distances of properties
from a linear feature centrepoint).
...
I understand I need to calculate a maximum permissible value (this will
be different for each grouping), my problem is how calculate this value
(exclude the values of the records considered in error) before I know
which records are in error!!
----------------------------------------------------------------------
To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put
"unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]

Reply via email to