POLSCI 391 Chapter Notes - Chapter 3: Durable Good, Markov Chain, Markov Model
Document Summary
Markov processes capture an entity or population that transitions among a finite set states according to fixed probabilities. Used to explore student engagement, political transitions, and the ranking of web pages, academic journals, or sports teams. Statistical equilibrium: entities keep moving around, but an cumulative statistic remains unchanged. A finite number of states (categories of the set of possibilities) Set of fixed transition probabilities for moving between states. Markov model can be applied to single entity or population. Single entity- can tell us the likelihood of that entity being in each state. Population- predicts the percentage of individuals in each state. In a markov model time unfolds in discrete periods, such as hours, days, months, or years each period, the model is in some state and then transitions to some state in the next period. Fixed set of states, fixed transition probabilities, the possibility of getting from any state to another through a series of transitions, and no simple loops.