In psychology, we increasingly encounter data that is nested. It is to the point now where any quantitative psychologist worth their salt must know how to analyze multilevel data. A common approach to multilevel modeling is the varying effects approach, where the relation between a predictor and an outcome variable is modeled both within clusters of data (e.g., observations within people, or children within schools) and across the sample as a whole.
A Markov Chain describes a sequence of states where the probability of transitioning from states depends only the current state. Markov chains are useful in a variety of computer science, mathematics, and probability contexts, also featuring prominently in Bayesian computation as Markov Chain Monte Carlo. Here, we’re going to look at a relatively simple breed of Markov chain and build up some intuition using simulations and animations (two of my favorite things).
Binomial probability is the relatively simple case of estimating the proportion of successes in a series of yes/no trials. The perennial example is estimating the proportion of heads in a series of coin flips where each trial is independent and has possibility of heads or tails. Because of its relative simplicity, the binomial case is a great place to start when learning about Bayesian analysis. In this post, I will provide a gentle introduction to Bayesian analysis using binomial probability as an example.
This past summer, I watched a brilliant lecture series by Richard McElreath on Bayesian statistics. It honestly changed my whole outlook on statistics, so I couldn’t recommend it more (plus, McElreath is an engaging instructor). One of the most compelling cases for using Bayesian statistics is with a collection of statistical tools called linear mixed models or multilevel/hierarchical models. It’s common that data are grouped or clustered in some way.