Hi list,
More general GIS analysis than pure MI
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).
e.g.
25
34
56
57
59
60
62
1189
1190
The distances in each group vary but are "grouped".
I wish to identify any records that are likely to be errors, the
criteria for an error is that they form outliers in this group of
values.
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!!
I have looked at sorting the group by value and performing a test on the
last and its preceeding record (eg if lastrec < 2*prevrec then Pass,
else Fail) which works if only one record is an outlier, however if two
or more records have similar values (all errors) this becomes less
suitable.
Any advice, suggestions welcomed
Mike Atkinson
----------------------------------------------------------------------
To unsubscribe from this list, send e-mail to [EMAIL PROTECTED] and put
"unsubscribe MAPINFO-L" in the message body, or contact [EMAIL PROTECTED]
