Summary: Statistics Ii

Read the summary and the most important questions on Statistics II

• Lecture 1

  This is a preview. There are 2 more flashcards available for chapter 27/01/2021
  Show more cards here

• P value and effect size?

  P value tells you nothing about the effect size or the value of the effect. 
  you should focus less on p value and more on the effect size (ES). 

• What tells the P value

  P value tell something bout the data, not anything about H0 being true or not. Tells how surprising is your data considering the H0?

• Transforming plot into log scale, why would you do this?

  To reduce variability

• Lecture 4

• How to get increase the power of your statistics test?

  By increasing the sample size, the power of your test will increase aswell. (not always, sample size should also not be to big -> so calculate sample size (see slide 49, 62))
  by inducing the variability, the power will increase aswell.

• Lecture 5 Bayesian statistics

  This is a preview. There are 3 more flashcards available for chapter 24/02/2021
  Show more cards here

• What is a Bayesian P-value?And the classical P-value?

  • Hypothesis given the data, how likely or data and H0 are.
    
  • the p-value is the probability of obtaining results at least as extreme as the observed results of a statistical hypothesis test, assuming that the null hypothesis is correct
    

• Uninformative prior (horizontal)

  You have no idea what is going to happen, gives no info -> than posterior is completely determined by you data (=likelihood)

• What defines a prior (expectation)?

  Depends on you, your expert knowledge about the topic (e.g. Drug) decides the prior. Not an wild guess but more like an educated guess.

• Computation of posterior is easiest for which scenarios ?

  • With a binomial- or normal-likelihood/distribution.
    
  • without, so with more complex models, you do a Markov Chain Monte Carlo (MCMC) procedure/simulation
    

• How is the H0 formulated in Bayesian statistics ?

  What evidence to my data give, given my hypothesis (effe nakijken)

• Algemeen notes

  This is a preview. There are 81 more flashcards available for chapter 13/03/2021
  Show more cards here

• What do P values tell you?

  Tells you how surprising your data are when H0 would be true.
  
  Much more interesting what evidence your data give to H0, reversed probability -> Bayesian statistics.