BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.

Hello,

 

Where can I find the way SAS Miner calculates the misclassifiation rate, MISC, in the scorecard node in Miner?

 

Thanks!

1 ACCEPTED SOLUTION

Accepted Solutions
DougWielenga
SAS Employee

The Misclassification rate can only be calculated once a prediction is made for a categorical target variable.   When building a model against a categorical target variable, SAS Enterprise Miner will calculate the probability of each level of the target.  The choice for the predicted level then depends on what you have used to choose a predicted target level.  By default, SAS Enterprise Miner will predict the most likely outcome for each observation and store in it a variable of the form I_< target variable name> ; however, SAS Enterprise Miner can also incorporate Decision weights which can are used along with the predicted probability of each level to calculate an expected value of each decision.  In this latter case, it chooses the most profitable (or least costly) decision and stores the resulting level in a variable of the form D_<target variable name>.      


Suppose you have a binary target variable Purchase which equals "1" or "0" for each observation.  SAS Enterprise Miner will compute 

 

    P_Purchase1 = probability that Purchase = "1"

    P_Purchase0 = probability that Purchase = "0"

    I_Purchase = most likely outcome based the greater of P_Purchase1 and P_Purchase0

 

but it might also compute 

 

     D_Purchase = most "valuable" outcome after computing the product of the probability of each outcome and the associated decision weight

 

when you have specified a decision profile and clickec the radio button for Yes under "Do you want to use the decisions" in the Decision Processing dialog available from the Input Data Source node or a Decisions node.  As a result, the answer to your question depends on whether you had specified and used decision weights or not.   

 

Hope this helps!

Doug

View solution in original post

2 REPLIES 2
M_Maldonado
Barite | Level 11

Hi Rogelio,

It looks like most fit statistics are defined only once on the Reference Help (press F1 when you are on Enterprise Miner to see the Reference Help). If they are defined on the Model Comparison node, you don't see them again on the Scorecard node, unless they are a specific result of the Scorecard node.

 

I hope this helps!

M

DougWielenga
SAS Employee

The Misclassification rate can only be calculated once a prediction is made for a categorical target variable.   When building a model against a categorical target variable, SAS Enterprise Miner will calculate the probability of each level of the target.  The choice for the predicted level then depends on what you have used to choose a predicted target level.  By default, SAS Enterprise Miner will predict the most likely outcome for each observation and store in it a variable of the form I_< target variable name> ; however, SAS Enterprise Miner can also incorporate Decision weights which can are used along with the predicted probability of each level to calculate an expected value of each decision.  In this latter case, it chooses the most profitable (or least costly) decision and stores the resulting level in a variable of the form D_<target variable name>.      


Suppose you have a binary target variable Purchase which equals "1" or "0" for each observation.  SAS Enterprise Miner will compute 

 

    P_Purchase1 = probability that Purchase = "1"

    P_Purchase0 = probability that Purchase = "0"

    I_Purchase = most likely outcome based the greater of P_Purchase1 and P_Purchase0

 

but it might also compute 

 

     D_Purchase = most "valuable" outcome after computing the product of the probability of each outcome and the associated decision weight

 

when you have specified a decision profile and clickec the radio button for Yes under "Do you want to use the decisions" in the Decision Processing dialog available from the Input Data Source node or a Decisions node.  As a result, the answer to your question depends on whether you had specified and used decision weights or not.   

 

Hope this helps!

Doug

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 7330 views
  • 0 likes
  • 3 in conversation