Many preprints and published papers report results of Monte Carlo experiments with too many ‘insignificant’ digits.
A benchmark of different approaches to calibrating proposals in a geostatistic model with geometric anisotropy.
Different strategies for the design of Markov chain Monte Carlo using random walk Metropolis proposals, Metropolis-adjusted Langevin algorithm. We consider a simple beta-binomial scenario with parameter transformations.
Inadequate statistical analysis explains patterns uncovered in a paper published in Nature on maximum age at death of humans.
This post examines the claim that the score statistic for the \(r\)-largest order statistics likelihood is irregular.
This short note shows that inverted max-stable vectors cannot have upper tail dependence.
A short proof that quadratic form of a Gaussian vector verify positive regression dependence when the entries of their covariance matrix are positive.
