cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
waleye
‎08-02-2016
  • 4 Posts
  • 1 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: A statistic question about model

by Calcite | Level 5 in SAS Data Science
‎07-20-2016 02:16 PM
‎07-20-2016 02:16 PM
hi ballardw  This is a good question. If %KX_Bene was used. Some small provider. Like Provider D. He has only 1 beneficiary, who is fraud bene. So, %KX_Bene is 100%, the highest. But, obviously, he is not the provider who we want to catch. If KX_Bene*%KX_Bene  was used, then the number of fraud_bene was taken into account.  KX_Bene*%KX_Bene  is like a "Weight" or "Expectation". Acctually, if KX_Bene*%KX_Bene  was used, the performance of the model become better. But, I cannot find the theory to support that.       ... View more

Re: A statistic question about model

by Calcite | Level 5 in SAS Data Science
‎07-20-2016 09:57 AM
‎07-20-2016 09:57 AM
Hi rayIII  Thank you for your response. For example. KX is the maximum payment has been reached. Provider Total_bene     KX_Bene  %KX_Bene    KX_Bene*%KX_Bene A            1000               600            0.6                360 B            100                 70              0.7                49 C            10                   8                0.8                6.4 D            1                     1                1.0                 1 Most of the model use %KX_Bene as detector. But I want to use "KX_Bene*%KX_Bene" as a detector. Do you think it is reseanonable? if so, is there any statistical theories to support that? Thanks. ... View more

Re: A statistic question about model

by Calcite | Level 5 in SAS Data Science
‎07-20-2016 09:52 AM
‎07-20-2016 09:52 AM
Hi Reeza  Thank you for your response. Yes, you are totally right. Most of the models use the amount or percentage as detector. But I want to try used Fraud*%Fraud as  detector. So do you think Fraud*%Fraud is reasonable?  and do you have a literature to support this? Thanks. ... View more

A statistic question about model

by Calcite | Level 5 in SAS Data Science
‎07-19-2016 10:31 PM
‎07-19-2016 10:31 PM
Provider Charge(Total) Fraud  %Fraud    Fraud*%Fraud A           1000               600        0.6           360 B            100               70           0.7            49 C            10                 8             0.8            6.4 D            1                   1             1.0            1 Which provider should be given alert? I am creating a model and setting up a threshold as above. Which provider should be given alert? If %Fraud is used, then provider D should be given alert. Obviously, it is unreasonable. Then, I want to use Fraud*%Fraud, the results are pretty reasonable. Question: 1)Are there any statistic theory about Fraud*%Fraud? If so, what is the theory? 2)If Fraud*%Fraud is unreasonable. Are there any better suggestion? Thanks in advance. ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎08-02-2016 05:17 PM
Likes given to
View All
Quick links