Statistical Models

The Statistical Inference Part discusses fundamental statistical inference methods, such as interval estimation and hypothesis testing. The methods can be classical/frequentist, Bayesian, parametric, nonparametric or resampling-based such as bootstrapping.

In this Part: Statistical Models, we focus on more sophisticated statistical methods including Analysis of Variance, Linear Regression, Bayesian Linear Regression, and Survival Analysis. Of course there are tons of other statistical models out there, but the models discussed in this book are basic but important ones because many models are based on those models, and improve their performance.

A statistical model is a mathematical model that describes the data generating process which is the core of inferential statistics. The methods discussed the Inference Part can be viewed as a very basic statistical model. A sophisticated statistical model not only does the inference about unknown population parameters but also make prediction about future or unseen observations or data, the core of modern statistical machine learning.