BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
nicopalacios
Calcite | Level 5

Hello everyone!

I have a problem. I'm doing a credit scoring model, and the cutoff of probability is 0.50. How should I do to change the cutoff?

1 ACCEPTED SOLUTION

Accepted Solutions
M_Maldonado
Barite | Level 11

The cutoff node is just for analysis, but it will not change the cutoff on your score code. Use it just to find the value of the cutoff.

Once you have found a good value, and you are ready to put your code into production, run a Score node to get the optimized score code and manually change the 0.5 for some other value. By manually change I mean that you will hardcode this value when you put the score code in the environment you are planning to use for scoring new observations.

Does that help?

Thanks,

View solution in original post

4 REPLIES 4
M_Maldonado
Barite | Level 11

There are a couple ways to do this.

First you determine a good cutoff value (for example you can use Event Precision equals Recall option as suggested in ).

Next, you use a Score node to generate the optimize score code that you would put into production. And you change the value directly on the score code.

Another alternative would be to use the score code as it is to generate the predicted probabilities and write your own code to assign predicted events based on a new cutoff for the posterior probability.

I hope this helps!

-Miguel

nicopalacios
Calcite | Level 5

Hi Miguel, thanks for your answer.

I can not find where to change the cutoff. I tried in modificarle the "User imput," but it turns out to see the "data exported" continues using 0.50 as the cutoff. And in the Score node can not find an option to change the cutoff. What would be the correct procedure? Scorecard > Cutoff > Score?

M_Maldonado
Barite | Level 11

The cutoff node is just for analysis, but it will not change the cutoff on your score code. Use it just to find the value of the cutoff.

Once you have found a good value, and you are ready to put your code into production, run a Score node to get the optimized score code and manually change the 0.5 for some other value. By manually change I mean that you will hardcode this value when you put the score code in the environment you are planning to use for scoring new observations.

Does that help?

Thanks,

nicopalacios
Calcite | Level 5

excellent!

thank you

sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

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
  • 4 replies
  • 2749 views
  • 3 likes
  • 2 in conversation