"Whether we and our politicians know it or not, Nature is party to all our deals and decisions, and she has more votes, a longer memory, and a sterner sense of justice than we do." — Wendell Berry
I’d love to see what this does in various kinds of regression. It may be possible to set up some kind of iterative regression scheme, where a normal regression with uniform weights is first done, and then the residuals are used to define a set of alternative weights via the Tukey Loss Function. Then the weighted regression is done, producing another set of residuals, and a new set of weights is defined. This should (eventually) settle down.
The Tukey loss function, also known as Tukey’s biweight function, is a loss function that is used in robust statistics. Tukey’s loss is similar to Huber loss in that it demonstrates quadratic behavior near the origin. However, it is even more insensitive to outliers because the loss incurred by large residuals is constant, rather than scaling linearly as it would for the Huber loss.
In the above, I use $latex r$ as the argument to the function to represent “residual”, while $latex c$ is a positive parameter that the user has to choose. A common choice of this parameter is $latex c = 4.685$: Reference 1 notes that this value results…
See https://wordpress.com/view/667-per-cm.net/ Retired data scientist and statistician. Now working projects in quantitative ecology and, specifically, phenology of Bryophyta and technical methods for their study.
Mike Bloomberg, 2020
He can get progress on climate done, has the means and experts to counter the Trump and Republican digital disinformation machine, and has the experience, knowledge, and depth of experience to achieve and unify.
GeoEnergy Math
Prof Paul Pukite’s Web site devoted to energy derived from geological and geophysical processes and categorized according to its originating source.
Quotes by Nikola Tesla
Quotes by Nikola Tesla, including some of others he greatly liked.
Higgs from AIR describing NAO and EA
Stephanie Higgs from AIR Worldwide gives a nice description of NAO and EA in the context of discussing “The Geographic Impact of Climate Signals on European Winter Storms”
Why "naive Bayes" is not Bayesian
Explains why the so-called “naive Bayes” classifier is not Bayesian. The setup is okay, but estimating probabilities by doing relative frequencies instead of using Dirichlet conjugate priors or integration strays from The Path.
All about ENSO, and lunar tides (Paul Pukite)
Historically, ENSO has been explained in terms of winds. But recently — and Dr Paul Pukite has insisted upon this for a long time — the oscillation of ENSO has been explained as a large-scale slosh due to lunar tidal forcing.
Logistic curves in market disruption
From DollarsPerBBL, about logistic or S-curves as models of product take-up rather than exponentials, with notes on EVs
Darren Wilkinson's introduction to ABC
Darren Wilkinson’s introduction to approximate Bayesian computation (“ABC”). See also his post about summary statistics for ABC https://darrenjw.wordpress.com/2013/09/01/summary-stats-for-abc/
Mathematics and Climate Research Network
The Mathematics and Climate Research Network (MCRN) engages mathematicians to collaborating on the cryosphere, conceptual model validation, data assimilation, the electric grid, food systems, nonsmooth systems, paleoclimate, resilience, tipping points.
HotWhopper: It's excellent.
Global warming and climate change. Eavesdropping on the deniosphere, its weird pseudo-science and crazy conspiracy whoppers.
CLIMATE ADAM
Previously from the Science news staff at the podcast of Nature (“Nature Podcast”), the journal, now on YouTube, encouraging climate action through climate comedy.
All Models Are Wrong
Dr Tamsin Edwards blog about uncertainty in science, and climate science
Isaac Held's blog
In the spirit of Ray Pierrehumbert’s “big ideas come from small models” in his textbook, PRINCIPLES OF PLANETARY CLIMATE, Dr Held presents quantitative essays regarding one feature or another of the Earth’s climate and weather system.
Climate Communication
Hassol, Somerville, Melillo, and Hussin site communicating climate to the public
Équiterre
Equiterre helps build a social movement by encouraging individuals, organizations and governments to make ecological and equitable choices, in a spirit of solidarity.
Anti—Anti-#ClimateEmergency
Whether to declare a climate emergency is debatable. But some critics have gone way overboard.
"Warming Slowdown?" (part 1 of 2)
The idea of a global warming slowdown or hiatus is critically examined, emphasizing the literature, the datasets, and means and methods for telling such. In two parts.
Climate change: Evidence and causes
A project of the UK Royal Society: (1) Answers to key questions, (2) evidence and causes, and (3) a short guide to climate science