STAT 2 Lecture Notes - Lecture 1: Statistical Inference, Biostatistics, Model Checking
Document Summary
Quizzes are at the end of 2nd and 6th weeks. Variation is the reason for statistics if everything were the same, it would be meaningless. Statistics can overlap in fields such as political science, economics/finance, medicine (biostatistics), linguistics, computer science (i. e. ai, machine learning), psychology, agriculture. Reality (variability and uncertainty) data model (theory of probability) . Second arrow: statistical inference: parameter estimation, model checking. Experiments vs. observational studies -- how data is obtained. For experiments, there is a clear distinction between the control group and the treatment group. Confounder refers to effect which may blur the effect of the treatment itself. Google definition: a situation in which a measure of association or relationship between exposure and outcome is distorted by the presence of another variable. Randomization is powerful because it distributes both known and unknown risk factors approximately equally across both the control and treatment groups.