Points of significance: Bayesian statistics

  title={Points of significance: Bayesian statistics},
  author={Jorge L{\'o}pez Puga and Martin Krzywinski and Naomi S. Altman},
  journal={Nature Methods},
Today's predictions are tomorrow's priors. One of the goals of statistics is to make inferences about population parameters from a limited set of observations. Last month, we showed how Bayes' theorem is used to update probability estimates as more data are collected 1. We used the example of identifying a coin as fair or biased based on the outcome of one or more tosses. This month, we introduce Bayesian inference by treating the degree of bias as a population parameter and using toss outcomes… Expand
[Bayesian analysis. Basic and practical concepts for its interpretation and use].
Three examples of hypothesis to be tested are shown: difference of means, correlation and association and the hypothesis with the highest degree of certainty at its update is the one that is accepted for decision making. Expand
Deciding on Null Hypotheses using P‐values or Bayesian alternatives: A simulation study
The results show that Bayesian factors are good complements for testing a hypothesis, and are suggested to be used together with classical and Bayesian statistics together instead of rejecting either of them. Expand
Applications of Bayesian Skyline Plots and Approximate Bayesian Computation for Human Demography
Bayesian skyline plots and approximate Bayesian computation as applied to human demography are reviewed and the basic mechanics of these methods are described in an attempt to demystify them. Expand
Towards better estimation of statistical significance when comparing evolutionary algorithms
This paper develops statistical tests that are good in answering typical questions coming from benchmarking of evolutionary algorithms, and is able to spot very small differences in the performance of algorithms while requiring computational budgets by an order of magnitude smaller compared to the existing approaches. Expand
Experimental Planning and Execution
These measures will avoid a GIGO (garbage in, garbage out) paradigm that no amount of postexperimental tinkering, including P -hacking, can objectively and legitimately correct. Expand
Experimental measurements of break-up reactions to study alpha clustering in carbon-12 and beryllium-9
Due to the high binding energy of the α-particle, this object can preform in heavier atomic nuclei. This work explores α-clustering in \(^9\)Be and \(^12\)C by measuring their nuclear break-up. ForExpand
Flash floods versus river floods – a comparison of psychological impacts and implications for precautionary behaviour
Abstract. River floods are among the most damaging natural hazards that frequently occur in Germany. Flooding causes high economic losses and impacts many residents. In 2016, several southern GermanExpand
Global glacier mass changes and their contributions to sea-level rise from 1961 to 2016
The largest collection so far of glaciological and geodetic observations suggests that glaciers contributed about 27 millimetres to sea-level rise from 1961 to 2016, at rates of ice loss that could see the disappearance of many glaciers this century. Expand
Visualisation Methods of Hierarchical Biological Data: A Survey and Review
This paper reviews the current state of visualization methods in a biomedical context and focuses on hierarchical biological data as a source for visualization, and gives a comprehensive review of current techniques. Expand
Study Design for Sequencing Studies
High-throughput sequencing has become the primary data acquisition method for many genomics studies and is being used more and more to address molecular biology questions. Expand