Licence & Copyright
From the Bottom of the Heap by Gavin L. Simpson is, unless otherwise expressly stated, licensed under a Creative Commons Attribution 3.0 Unported License.
Permissions beyond the scope of this license may be available at ucfagls.wordpress.com.Gavin Simpson on Twitter
- @Jowanza if things were working before, why upgrade R until your workflow was secure? #rstats 6 hours ago
- @sjGoring Over several 1-hour sessions. Not really lectures - prefer more hands on with slides to keep things moving. 8 hours ago
- @sjGoring After that hook, I move on to domain-specific topics to get people doing things whilst learning R. Ramp up R level as you progress 9 hours ago
- @sjGoring the plotting focus is to illustrate how a few simple commands gets you a publication quality figure etc. 9 hours ago
- @sjGoring I've usually focussed on some basic R syntax/usage, then covered data import, and then got stuck into graphics/plotting 9 hours ago
- @sjGoring these are ecology undergrads or grad students or beyond? 9 hours ago
- @sjGoring to whom will you be teaching? (what level, what subject areas?) 9 hours ago
- RT @figshare: figshare launches Publisher portals figshare.com/blog/figshare_… #PLOS: plos.figshare.com #F1000Research: http://t.co/KHwYuLdk… 10 hours ago
- RT @KirkwoodLab: Insightful post by @mommiologist: The great escape: charting a career outside of academia wp.me/p2wMuW-1Ci via @Dyna… 1 day ago
- @l_marazzi Nope, hopefully attending the ESA meeting in Minneapolis around the same time though. 1 day ago
May 2013 M T W T F S S « Jan 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Meta
Tag Archives: overfitting
What’s wrong with LOESS for palaeo data?
Locally weighted scatterplot smoothing (LOWESS) or local regression (LOESS) is widely used to highlight “signal” in variables from stratigraphic sequences. It is a user-friendly way of fitting a local model that derives its form from the data themselves rather than having … Continue reading
Posted in Palaeoecology, Palaeolimnology, R, Science, Time series
Tagged autocorrelation, cross-validation, GCV, LOESS, overfitting
4 Comments
