- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
I know that SAS creates ROC curve and calculate AUROC for free when I run PROC LOGISTIC.
According to this blog, Area Under Precision-Recall Curve is more appropriate in quantifying discrimination power of a model than AUROC.
https://machinelearningmastery.com/roc-curves-and-precision-recall-curves-for-classification-in-pyth...
I tried searching online but could not find any instructions for plotting the curve or calculating AUPRC in base SAS (with SAS/STAT).
Is there a built-in function for it? If not, how can we implement this?
I am using SAS 9.4M4 with SAS/STAT 14.2
Thank you!
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Looks like this would be a roll your own unfortunately.
You can capture the output in a data set, do the calculations in a data step and then plot it using SGPLOT.
@clarkchong1 wrote:
I know that SAS creates ROC curve and calculate AUROC for free when I run PROC LOGISTIC.
According to this blog, Area Under Precision-Recall Curve is more appropriate in quantifying discrimination power of a model than AUROC.
https://machinelearningmastery.com/roc-curves-and-precision-recall-curves-for-classification-in-pyth...I tried searching online but could not find any instructions for plotting the curve or calculating AUPRC in base SAS (with SAS/STAT).
Is there a built-in function for it? If not, how can we implement this?
I am using SAS 9.4M4 with SAS/STAT 14.2
Thank you!