MSc Univariate Stats

Home

Publications

research

teaching

OpenMx

This is the content for my section of the MSc-Level course R course (course itself hosted on LEARN)

The degree program (DRPS) information on the course is here.

Week 3: Refresher in Inferential Statistics Hypothesis testing

In this lecture, we'll do a basic refresher in inferential statistics and hypothesis testing.

• Powerpoint slides are here
• Concepts include False positive (Type I) and the Type-1 error rate or p-value (α)
• Also well as false negatives (Type II) errors, the type II error rate (β)
• The critical concept of Power (1 − β) will be discussed (and discussed) in the context of how we efficiently discover the truth and avoid error.
• We'll also cover confidence intervals (CI95), and effect sizes (𝛽) and their strengths and limitations
• We'll discuss correlation (an unresolved causal nexus)
• Comparing proportions (χ²) with Ronald Fisher's Lady who drinks Tea
• We'll talk about Non-parametric statistics if we get time.

Some notes based on previous student questions appear below

• why shouldn't i make my continuous scales (like BMI) into categorical?
• good guide to linear modelling
• http://norvig.com/experiment-design.html
• http://norvig.com/prayer.html
• http://brenocon.com/blog/2011/05/log-normal-and-logistic-normal-terminology