‘Estimation and inference’ shows how statistics can be used to make statements about unobserved values. Point estimations can be made in a variety of ways, for instance the maximum likelihood approach, least squares estimation, posterior distributions, and the Bayesian approach. The reliability of each estimate can be assessed through its bias and mean squared error. Interval estimations give a range of values which we can be confident contains the true value (in Bayesian methods the credibility interval is calculated). Statistics can be used to test hypotheses, either by comparing a null and alternative hypothesis, or by accepting or rejecting a null hypothesis based on p-values.