Subscribe to DSC Newsletter

I am a stats newbie. A geoscientist with only undergrad stats training. Doing “pure” prediction, no clustering. Have generated a predictive algorithm using several different techniques. Best regression coefficient was 0.85 with K-NN. I expected to have to input geospatial data along with the data for the predictors (6) and predicted.

Can K-NN work like this or is it’s algorithm and predictions 100% spurious ?

Thanks. 

Views: 186

Reply to This

Replies to This Discussion

Hi Randolph, you might want to explore kriging, as this is a popular tool for geoscientists. The book from Noel Cressie also addresses this topic and much more in the context of spatial statistics, see here

Thank you again Vincent. I will follow-up. 

Vincent Granville said:

Hi Randolph, you might want to explore kriging, as this is a popular tool for geoscientists. The book from Noel Cressie also addresses this topic and much more in the context of spatial statistics, see here

Reply to Discussion

RSS

Follow Us

Videos

  • Add Videos
  • View All

Resources

© 2018   Data Science Central   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service