General information
Category:
Coding, syntax & commands
If I interpret your question correct, I'd generate a frequency table in vfp as Naomi did with cte
Select ID, count(*) as cnt from Mytabs && perhaps order by cnt
and feed that data first to a statistics program. Either a regression function can be fit or perhaps a cluster analysis can group the data.
Prearranging the # of bins is more likely to be incorrect
>Hello,
>
>I have a table with 1,500 records and with one column: ID, comprised of numeric data ranging from 1 - 1,591. As the
>numbers get higher, the frequency of their appearance increases. But what's important, is not so much any given value
>but how many numbers appear within a given range. And yes, I am trying to figure out a cutoff point that will separate
>this table into the numbers that appear with the highest frequency (they happen to be larger numbers, in this table)
>and numbers that appear less frequently (they happen to be smaller numbers.)
>So, I want to know if certain values appear more often than others, kind of like a heat map, except the response would
>be in a tabular form rather than in graphic.. Is there a way to use an SQL statement to do this? Can I preselect
>the number of ranges and their size, for example: break the table up into 20 ranges. But I can't know
>which values each of the 20 ranges will have in advance? I know it's a vague question but I wondered
>what you all might say as this kind of question is coming up a lot.
>
>Thank you,
>Steve
Previous
Reply
View the map of this thread
View the map of this thread starting from this message only
View all messages of this thread
View all messages of this thread starting from this message only