Frequentist statistics considers probability as the frequency of events over a large number of experiments. It is based on the assumption that the parameters generating the data are fixed and unchanging and that repeated sampling will cause the parameter estimates to get closer and closer to the true values.

In contrast, Bayesian statistics views probability as a subjective probability based on existing knowledge and experience. It uses Bayes' theorem to update the probabilities of hypotheses or parameters depending on existing data. Bayesian statistics also take parameter uncertainty into account.