Click on our Secret Library of Evidence ------>

    BANKILEAKS Secret Library

Loan Application Forms (LAF's)  

    Bank Emails to Brokers  

    Then Click on 'VIEW NOTEBOOK'

Join us on facebook

facebook3           facebook2 


What BFCSA Does...

BFCSA investigates fraud involving lenders, spruikers and financial planners worldwide.  Full Doc, Low Doc, No Doc loans, Lines of Credit and Buffer loans appear to be normal profit making financial products, however, these loans are set to implode within seven years.  For the past two decades, Ms Brailey, President of BFCSA (Inc), has been a tireless campaigner, championing the cause of older and low income people around the Globe who have fallen victim to banking and finance scams.  She has found that people of all ages are being targeted by Bankers offering faulty lending products. BFCSA warn that anyone who has signed up for one of these financial products, is in grave danger of losing their home.


Articles View Hits

Whistleblowers' Corner!

To all mortgage brokers, BDMs and loan approval officers! 
Pls Call Denise: 0401 642 344 

"Confidentiality is assured."

Cartoon Corner

Lighten your load today and "Laugh all the way to the bank!"

Denise Brailey

Led by award-winning consumer advocate Denise Brailey, BFCSA (Inc) are a group of people who are concerned about the appalling growth of Loan Fraud around the world. BFCSA (Inc) is a not for profit organisation in the spirit of global community concern and justice.

Click on the Cluster Map.

  • Home
    Home This is where you can find all the blog posts throughout the site.
  • Categories
    Categories Displays a list of categories from this blog.
  • Bloggers
    Bloggers Search for your favorite blogger from this site.
  • Login
    Login Login form

Automated Valuation Models reduce fraud risk for QBE LMI

  • Font size: Larger Smaller
  • Hits: 1518
  • Print
Automated Valuation Models reduce fraud risk for QBE LMI
March 2010
Executive summary
QBE LMI estimates that fraudulent and overstated property valuations have cost the Australian banking and finance industry tens-of-millions of dollars over the past decade. And with the increase in property values over that same period masking the full extent of the problem, obtaining accurate valuations of the properties used as security for mortgage lending is clearly one of the industry’s biggest challenges.  Automated Valuation Models (AVMs) have been available for many years but were perceived to be inaccurate and unreliable. However, new entrants in the Australian AVM market and recent enhancements in technology were the catalyst for QBE LMI to engage Quantium to evaluate the potential application of AVMs.
QBE LMI concluded that the current performance of AVMs was inadequate to be considered a commercially viable alternative to traditional valuations for the high-risk end of the mortgage lending and lenders’ mortgage insurance market...........QBE LMI, after conducting an extensive study, believes it has made a significant step forward to finding a solution to this problem: Automated Valuation Models also known as AVMs...........
Figure 1 – All AVM results plotted against ‘True’ value
Source: Quantium
Page 4
If the AVM and True Value were the same in every instance, all points would be plotted on the orange line that runs diagonally from bottom-left to top-right of the graph. Clearly they were not!........If all the AVM results were within ±20% of the True value, then all the points would be plotted within the upper boundary line +20% and the lower boundary line -20%. Again, clearly many of the results were outside of these boundaries.  As Figure 1 clearly shows, there were a number of outliers and the error rate of some of these outliers was quite significant........
If the underwriter and lender can not resolve these concerns, QBE LMI may decline the application for lenders’ mortgage insurance..........
Attachments area
Preview attachment LMI whitepaper AVMs reduce fraud risk ofr QBE LMI March 2010.pdf
LMI whitepaper AVMs reduce fraud risk ofr QBE LMI March 2010.pdf
900 KB
Last modified on
Rate this blog entry:


  • No comments made yet. Be the first to submit a comment

Leave your comment

Guest Monday, 28 September 2020