LAE

Welcome to the LAE community!  Please feel free to start a discussion in the discussion tab or join in a conversation.

Discussions

Members

Resources

Events

 View Only
  • 1.  Benford's law in LAE

    Employee
    Posted 11-06-2012 02:06

    Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

    Originally posted by: erno

    Hi, does someone have a solution to get Benford's law analysis done in LAE? (http://en.wikipedia.org/wiki/Benford%27s_law)

    This is somewhat standard feature in many data analysis products such as Caseware IDEA and ACL, and in my opinion would be very useful to have in LAE, too.

    Thanks!


  • 2.  RE: Benford's law in LAE

    Employee
    Posted 11-07-2012 00:03

    Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

    Originally posted by: mermacora

    Hi,
    It's relatively simple to implement the Benford's law logic in LAE.
    Some time ago I developed a couple of library nodes which perform simple 1 and 2 digit Benford�s law analysis on a set of currency amounts.
    A graph and some sample data to drive it are attached.
    The graph includes two library nodes:
    • 1-Digit Benford � extracts the 1st (left-most) digit of an input field and calculates the % distribution of this across the set of amount values. The Bedford probability is also calculated using the standard log10(1 + 1/n) formula and the deviation of the digit distribution from Benford is output.
    • 2-Digit Benford � as above but uses left-most 2 digits of an input field.

    Let me know if you have any questions about this.

    Regards,
    Mario Ermacora
    Attachments:
    Benford_Analysis.zip


  • 3.  RE: Benford's law in LAE

    Employee
    Posted 11-07-2012 00:34

    Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

    Originally posted by: erno

    This was exactly what I was looking for, thank you so much!