The Inference of Hypothesis Testing

This product has no pricing plans and so is not available for purchase

• 1

  The inference landscape

  • Hypothesis Testing, Bayesian Statistics, Akaike Information Criterion Free Preview
  • MMD, MML, Machine Learning, Causal Inference
• 2

  NHST overview

  • Historical origins and motivation
  • Gold standard of inference in 20th century
  • Reproducibility crisis and logical flaws emerge
  • NHST variations and modifications
  • Current status
• 3

  When to use NHST

  • Types of hypotheses and conditions
  • Experimental designs
  • When not to use NHST
  • Setting up a hypothesis and an experiment in Hypothesizen
• 4

  NHST's inferential step

  • Either we've discovered causation or been extremely unlucky
  • Null and Alternative Hypotheses
  • Gather the data
  • Find the p-value or critical value
  • Accept or reject the Hypothesis
• 5

  NHST's inferential step in Hypothesizen

  • All steps are automatically inferred from the experiment's description and Hypothesizen's AI
  • Hypothesizen's reasoning can be inspected and altered
  • Hypothesizen's coupled graphs reveals NHST's identical logic used in every test
  • Hypotheizen's coupled parameters allow new sensitivity analyses
  • Hypothesizen's pedagogical adjuncts allow new understandings of NHST's scope
  • Hypothesizen helps you to either understand or perform NHST tests more intuitiviely
• 6

  NHST as a fundamentally flawed methodology

  • A century of misunderstandings and misconceptions
  • Twenty NHST flaws that define a reproducibility crisis
  • Mitigation measures
  • Efforts to kill off NHST
• 7

  Hypothesizen as methodological segue

  • Using Hypothesizen to highlight reasoning flaws
  • Using Hypothesizen to understand past NHST experiments
  • Using Hypothesizen in rare scenarios where NHST still makes some sense
  • Using Hypothesizen to show NHST as a special case of Ockham's razor
  • Using Hypothesizen to transit from NHST to AIC, BIC, MLD or machine learning