HomeLatest ThreadsGreatest ThreadsForums & GroupsMy SubscriptionsMy Posts
DU Home » Latest Threads » Forums & Groups » Retired » Retired Forums » 2016 Postmortem (Forum) » Poll Averages, people. I... » Reply #8

Response to CrowCityDem (Reply #1)

Mon Jul 25, 2016, 11:59 AM

8. Let's be fair to Nate. His model is more sensitive to some kinds of swings than others.

Frankly I think he overfits his models a bit. Also, he uses mean values in his polls, which are more sensitive to outliers. Sam Wang at the Princeton Election Consortium uses a median-based approach instead which tends to disregard outliers.

But Nate isn't cheating. He is just reporting that outcomes his models predict.

Reply to this post

Back to OP Alert abuse Link to post in-thread

Always highlight: 10 newest replies | Replies posted after I mark a forum
Replies to this discussion thread
Please login to view edit histories.