# Monthly Archives: August 2019

## Errors-in-variables, or not …

One can see from this new arXival that errors-in-variables models have not yet become widely known within astronomy yet, though astronomers are trying to find ways to deal with this sort of modelling scenario. The errors-in-variables regression problem occurs when … Continue reading

## Prescience

I mentioned the concept of a Korean remake yesterday and today on the arXiv one appeared, almost as if I predicted it: this time it’s H0liCow Redux via GPs. It’s quite a cottage industry: obtain new cosmological data, whack a … Continue reading

## Some more abuse of Gaussian processes …

Two arXivals today show us ways to mis-use or mis-apply Gaussian processes. The first is another in this crazy series of papers trying to use GPs as extrapolators to estimate from datapoints. Yes, GPs can kind of extrapolate, but it’s … Continue reading

## Gaussian process with 10 datapoints

This arXival (accepted in Icarus) proposes to offer insights into the relative support for a periodic (seasonal) signal in Martian methane levels with respect to an aperiodic alternative. The method adopted is to fit various flavours (kernels) of a Gaussian … Continue reading

## Free-knot linear splines: Value of asymptotic theory

Another recent arXival proposes to use a local linear regression as a type of ‘non-parametric’ Bayesian model, although this time the location of the break points is learnt during fitting, which creates a free-knot linear spline model. The application here … Continue reading

## BIC is trash

Just a quick reminder that the BIC is not really an approximation to anything meaningful. Definitely not what’s suggested in this arXival.